Statistics Done Wrong: How to Avoid Common Stats Errors w/ Dr. Debbie Berebichez @Debbiebere (Episode 10)#DataTalk

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 PlayStitcherSoundCloud and Spotify.

In this #DataTalk, we had a chance to talk with Dr. Debbie Berebichez about the ways to spot and avoid common statistics errors. Make sure to follow our Deborah on TwitterLinkedIn, and check out her website: Science with Debbie

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 suggest future data science topics or guests, please contact Mike Delgado.

Here is a full transcript of the interview:

Mike Delgado: Hello and welcome to Experience Weekly Data Talk, a show featuring some of the smartest people working in data science. Today we’re excited to feature Dr. Debbie Berebichez, who is the chief data scientist at Metis, as well as a physicist and TV host for Discovery Channel’s Outrageous Acts of Science. Debbie has a PHD in physics from Stanford, and completed two post-doctoral fellowships at Columbia university and New York University. Her work in science has been featured all over, including the Wall Street Journal, Oprah, Dr. Oz, TED Talks, Wired, and dozens of other top, respected publications.

Aside from her stellar work in data science and physics, she’s also received the top Latina Tech Blogger award from the Association of Latinos in Social Media, and she’s the first Mexican woman to graduate with a PHD in physics from Stanford. Debbie, thank you so much for being our guest of Data Talk today.

Debbie Berebich: Thank you so much Mike. I’m so honored to be here.

Mike Delgado: Debbie I’m wondering if we can start by you of sharing your background in math’s, and science, and what led you into your path to becoming a data scientist.

Debbie Berebich: Yeah, of course. I grew up in Mexico City in a conservative community that discouraged women from pursuing a career in STEM. And from a very young age I was extremely curious about math, and the universe, and nature, and physics. I loved it, the little that I saw in school. But my teachers in school told me that that was not a career that was appropriate for a girl, and that I better pick something more feminine. And it was no better at home, or friends. My friends in school told me when I was teenager that in order to succeed in physics you had to be a genius. And I knew that I wasn’t a genius.

And then there were my parents at home. My mother is from Guatemala and she didn’t go to college. She said, “Oh, don’t tell boys that you like math because you’ll intimidate them, and you probably won’t be able to get married.” Almost happened. No, I’m kidding. But It was a very direct biased environment that made me hide my love for physics and math over the years. When it came time to go to college, I chose Philosophy because I was so inquisitive, I still wanted to ask questions. This was a more acceptable career path, I said, “well I’ll still be able to ask questions about nature and the universe just using a simpler method.”

But all throughout the two years I was doing philosophy, I realized that I couldn’t quiet that intense love and curiosity I had for physics.  In the middle of my B.A. in Mexico City, I applied to schools in the US because I found out that in the US you could do two majors. I could finish philosophy but also do physics. I applied behind my parent’s back and all that. I remember thinking what a rascal I was being. I would read books in the library of obscure physicists like Tycho Brahe, who was a Danish astronomer and he was locked in a tower. He was this very anti-social man who lost his nose in a duel. Kind of a nasty person. I thought “okay, maybe I am like him.” He’s my hero, so to speak, and I’m going to be anti-social. I’ve been locked up in a lab for all my life, but I’ll have my treasure, which was meticulously observing something about nature. And so here I had my sort of quiet secret heroes. And I read a lot about them, so when I applied to schools there was one problem, which was financial.

As a lot of students from overseas know, the universities in Mexico cost maybe eight times less than the ones in the US. So, I thought there’s no way I’m going to be able to make my dream come true. But all of a sudden, I had very good grades, and I had great letters of recommendation. Brandeis University, a small school, a wonderful one in Massachusetts, sent me a letter saying “You have good grades, we’d love to include you in this scholarship, if you write another essay and if you pass this test.” So, I was very fortunate to have won a Wien scholarship, which enabled me to transfer in the middle of my BA and I was going to start in my junior year at Brandeis.

I was super excited, and I flew in the middle of the winter. I had not experienced the snow. I came to Boston and I started studying philosophy and I had the courage in my first semester to take a very general astronomy course and I was fascinated by the stars and I befriended the teaching assistant in this course. His name is Roopesh, and he was a student from India who was doing his PhD at Brandeis. And Roopesh said that I wasn’t a typical student that just wanted to hear how to solve the homework problems and all that. He said to me “You have talent, but most importantly you have perseverance. And you should give it a try.” And he was the first person to really believe in me.

So, one day we’re walking in Harvard square, we’re sitting under a tree and I have tears in my eyes, and I look at Roopesh and I say “You know what, I don’t want to die without trying. I don’t want to die without trying to do physics.” So we called the head of the department at Brandeis, his advisor. He calls me to the office and he hands me a book. He says “You know what, somebody else did this in the past. His name is Ed Witten. Many years ago, he switched from history to physics in only two years.” Cause I only had 2 years to cram the entire physics major. Which normally is four years. And I said, “What? Ed Witten?” For those of you who know, he’s the father of string theory. So, he’s clearly a genius. I was, thinking he’s pulling my leg. There’s no way I could ever do this. In that moment, Roopesh decided to change my life. He became my mentor. I had two months because what Dr Wardle said is “Here’s a book called Div Grad and Curl; which is vector calculus in three dimensions. If by the end of the summer, in two months, you’re able to master this material, we’ll let you skip from the first two years of the physics major, so you can complete it in only two years.”

And I tell you this beautiful story because Roopesh devoted every single day of that summer tutoring me, mentoring me, helping me, pretty much cram two years into two months. And I always wanted to pay him for all his efforts and his mentoring me and tutoring me; and Roopesh said to me “You can’t pay me back, because when I was growing up in Darjeeling, in India, in this mountain town, there was an old man who use to climb up to this town and teach me and my sisters music, the tabla, English and mathematics. And whenever we wanted to pay this old man back, he said “No, the only way you could ever pay me back, is if you do this with someone else in the world.” And that’s how Roopesh passed the torch to me and created my mission in life, which is to inspire and encourage and help other minority, minorities especially women, who like myself feel attracted to STEM, science, engineering, technology and math, but who for some reason feel that they cannot achieve their dreams.

And so, fast forward, I was able to finish the physics and the philosophy degrees at Brandeis with highest honors. I then went back to Mexico and everyone said, “Okay, your adventure is done. Now stay here, get married and follow your normal life path.” But I was still very hungry for knowledge. I went to my advisor in Mexico and I told them, “Jose, Luis, I want to go back to the US because I want to pursue a PhD and the kind of physics that I’d like to do, we don’t have the resources for it in Mexico.” He says, “Okay, I understand. So, where did you apply?” I said, “Look, I sent emails, I know it’s late in the game. I sent emails to a bunch of people. But there’s one guy I liked. He’s at Stanford and he’s studying DNA, you know the physics of DNA single strands of DNA.” He said “Oh, who’s that?” And I said “his name is Steve Chu.”  And my advisor’s jaw dropped.He sa id “Steve Chu?”  I said “Yes, why?” He said, “Did you know that he just won the Noble prize in Physics a few months ago?” This was in 97.

And I said “Oh my God.” I’m not at all a shy person, but had I known that fact, I probably wouldn’t have written to him. Or I wouldn’t have been so casual as like “hey, what are you up to?” You know? And I was very fortunate. He invited me to work directly with him in his lab and then I changed. I became a theoretical physicist, but six years later I became the first Mexican woman to get a PhD in physics from Stanford. And I realized that with that privilege came responsibility to spread the word. I work in physics. I did two post-doctoral fellowships in New York in applied math and physics. And then I decided that academia was a little bit too isolating for me. I wanted to be more social. I get a chance to educate people and to explain to them these complex analysis concepts that I so loved and how to incentivize people to be critical thinkers. To me, the best gift you can give to anyone is that.

I worked for some time in Wall Street, which a lot of physicists do after physics creating risk models; using the math and statistics that I knew from physics and then I discovered that what I was doing was part of what people were calling data science, this new field. And so, I said “that sounds really interesting” and I don’t care about investing and just working with money all the time. That’s not my interest. I want to solve other more important problems in the world. So, I searched and searched and found my first job as a traditional data scientist at Thought Works, which is a consulting company, and finally I ended up moving to Metis, where I am now; which is a data science training company. And I love everything about using data to discover insights about the world and different companies and whatever happens around us.

Mike Delgado: Hearing your path with data science and your love of philosophy, which I think helped you in physics, right? Because philosophy’s all about how to think, logic; which plays a role into what you’re doing with the other science. I also love seeing how your passion led you into a place now, where you’re training and equipping others through Metis and classes. Especially with bringing in more women or minorities to get exposure to data science and that’s been one of the major problems that we’re seeing right now across the board when you look at a lot of companies and data scientists. A lot of them tend to be men. And we need to see more women involved in data science as well as minorities, because having different perspectives is so crucial, right?

Debbie Berebich: Absolutely, Mike. To that end, a group of seven executive women created the first data science curriculum for high school girls of underserved backgrounds and we did that three years ago. Since then we’ve deployed it in several high schools across the US through Girls Inc. and it’s been amazing. One of them was my mentee here at Metis last summer and she’s 16 years old and she was able to complete a project doing analysis of who takes our boot camps and you know what kind of success metrics we can measure to know who’s really thriving with our boot camps and what do we need to improve. It’s fascinating to see how you can change people’s lives, and especially women because that’s sort of the core of my mission.

Mike Delgado: Yeah, that’s beautiful. And just thinking about five years from now, ten years from now, as kids are getting out of high school moving into college, and we see all these predictions right about the importance of data science and business. No matter what role you’re in as we have more involvement with AI and machine learning at companies, everyone’s going to need to have a background at some point in data science. The work that you’re doing is going to be pivotal for the youth.

I think you are fascinating because as a child you were already hungering and craving physics, the hiding the books, which I love. But what about the kids who feel they’re not strong at math and I don’t know if data science is the right fit for me. What’s your message for those kids?

Debbie Berebich: Yeah, nobody is not strong at math. When I started going for the physics degree and I started my junior year, I’ll tell you that my algebra, A + B, all that squared, was rusty. I remembered very little from high school. People, humans, we are mathematical in our intuition, it just takes perseverance and a lot of hard work. Nothing of value is going to come easy in life; but if you put your minds to it and if you put effort, anyone can learn these topics. I always say that if I could do it, anyone can do it.

Though I have a 10-month-old baby now, and I can tell you that even now I see the way she moves around the world, she intuits certain things with her body about momentum and different physics things and I think it’s important as parents, especially, if you just cultivate the, what Feynman a very famous physicist use to say, “the pleasure of finding things out.” Nowadays, Mike, I think a lot of people focus on, especially in the US, multiple choice tests, on the correct answer. Science is not about that. Science, and the history of science, is full of failures and negative results. It’s about having pleasure in the path of discovery. And that’s what you should teach your children.

Whenever they come up with a question, as silly as it may seem “like why is the sky blue, mom?” Or “why does cheese melt?” And “why are there seasons around us?” All those questions, don’t just google them and don’t just give them the answer. Actually go through the process of finding the answer with them. Have a go at it and say “Oh wait, the sky is blue. Could it be because there’s a reflector that is blue? Or there are cones in our eyes, well that doesn’t sound reasonable, cause why would we see other colors?” Etc, etc. So, there are ways that you can take people through the path of learning, and if children learn to enjoy that and they’re not criticized for asking questions, it’s very likely that they’ll enjoy doing math, which is the most important thing.

Now, for all of us that are listening, we’re probably not children anymore. Those of you, I tell you, I have people in their 40’s, people who want to switch careers in our boot camps. Or even people who write to me privately and they’re like “you know I have my kids already and I want to switch careers.” Or people that come to our boot camp after doing marketing or something that didn’t have a lot of quantitative background. We teach them the basics. And yes, they may not be able to become a professor of math in five years, but they can learn enough math to become proficient data scientists.  I guarantee it.

Mike Delgado: That’s beautiful and I think it’s essential. No matter what role you’re at right now in business, whatever you’re working in, getting a background, taking an online course through Metis, or finding some program where you can begin to learn about data science, will bring value in the future. Getting a head start now is going to be very, very important for the future.

Debbie Berebich: Absolutely. Yes.

Mike Delgado: I love how you’re talking about as a parent. I just read through Everybody Lies by Seth Stephens and he talks about how we have this hidden bias as parents. Looked at Google searches between how people search for things about their sons versus daughters, and one of the unfortunate hidden biases is that parents were more likely to type in “is my son gifted?” And for girls it was “is my daughter overweight?” Those are the two parallel problems, the hidden bias.

Debbie Berebich: Yeah, it’s sad and you see it. And I must fight direct bias because I was literally told that I wouldn’t be able to do it. And sometimes that bias hurts more, but it’s easier to fight. The one that is more complex is the subtle bias, right, kind of hidden things in the media. Looking at TV commercials where it’s only the men doing mechanics and fixing the cars. Where you go out to a restaurant and it’s the man at the table that calculates the tip, or what not.  All those subtle things send messages to kids today that what matters for girls is very different than what matters for boys. And by the way I saw somebody in the audience ask for the Metis link, and you guys are totally welcome to look. We have online courses, we have corporate training, we have boot camps in four cities and we accept international students. The link for Metis is

Mike Delgado: I’m going to put that on the screen, That’s the place where you can go to find out all about the different online courses that are available. I highly recommend visiting.

Mike Delgado: And just hearing your background and your passion, I think it gets a lot of people excited about taking this on. Maybe even those who never thought about studying data science think “hey, this is something that I might want to do.”

Debbie Berebich: Yeah, I love helping in that way.

Mike Delgado: So, just briefly today, I want to talk specifically about statistics, which is a big part of data science. I was talking to my wife earlier, I just picked up a book on statistics, and she asked “why are you reading that?” I’m like “Well, I’m talking to somebody about statistics this next week.” And she said “well I took a credit course in college.” I’m like “But I need a refresher. I don’t remember a thing about statistics.” So, I know unfortunately there’s a lot of flawed statistical data that’s published through the media. Just recently with the last election, false news, false stats being spread. I’m curious about what are some of the myths or the pet peeves that you have with statistics, things that you see published, that drive you crazy.

Debbie Berebich: Yeah, I mean there are tons of things that drive people crazy, or should drive people crazy. In terms of myths I just this past week I’m experiencing different thoughts due to the devastating earthquakes in Mexico, as some of you know. For one example is that myth that has been propagated that the safest place in a house is under a doorway. So, during an earthquake, you should all go there. And that’s a complete myth and it started with a pretty small bias, which is a common example of a statistical failure. What happened in the late 1800’s is that there were adobe homes in California that during an earthquake the home fell, but the doorway was standing. And those images created this myth that “Hey, we should go to the doorway,” when in fact modern homes is quite the opposite. You can go to the door way, and the doors can swivel and hit you while the earthquake is going on, so that’s not something you should do. Instead, you should take cover under a metal table or something like sturdy and hold it with the other hand. But that’s a myth and it shows you how sometimes small samples can propagate myths for hundreds of years.

Other ones are, in the medical field for example, correlation is not causation. That’s a big one. People tend to think that because two things happen at the same time, one may be causing the other. That could not be further from the truth. There’s a beautiful website called, that shows you things that like a perfect correlation of the consumption of margarine is perfectly correlated in some time window with the divorce rate in Maine. There’s no way that one causes the other, yet because they’re correlated people tend to think that causation is there. That’s a silly example but it happens all the time when people say “I get colds all the time, but whenever I drink water with vinegar in the mornings, I don’t get colds.” And you know it may be just by chance you were already getting cured from your cold, and you know they happen at the same time, but your mind is biased or putting them in causal relationship so you’re going to tell everybody that vinegar with water cures colds.

That happens in medicine a lot, or not in real medicine hopefully but in these kinds of cold remedies, and unless you have randomized double blind placebo controlled experiment, you can’t assess that with this degree of certainty you know this medication or this method can cure such an illness. So, you must be very careful when you read the news and you read what the media is saying because people often have good intentions and simple mistakes that people make in statistics, people with the skills to do manipulate graphs, visual graphs and statistics to make you believe something that is not there.

Mike Delgado: Yeah I know just from working with Excel, playing with a chart and making it longer, longer, longer, all of a sudden the line looks lot more important, right?

Debbie Berebich: Of course, you can make something that just increased by 4% look like it increased four times because you’re pulling the axis from zero to there. There are tons of things like that.

Mike Delgado: And it’s funny you mention, some of those home remedy myths, I remember my grandma always talked about, and she swore by this, a concoction she would make. Apple cider vinegar, garlic and honey all mixed together and that was supposed to be the cure all for pretty much everything.

Debbie Berebich: In Mexico, we always say tequila cures your cold, if it doesn’t cure it, at least you can forget about it.

Mike Delgado: When I was little, whenever I had a cough, my dad would give me like a little spoonful of tequila which would burn my throat.  It was supposed to get rid of the mucus. That was the key.

I’m glad you shared some of those statistical errors about sample sizes and correlation does not equal causation and where is a sample coming from. Cause that can be a huge bias. There’s these powerful stats software’s that are available to help scientists reduce mathematical errors, but sometimes those errors that are being made aren’t from the software. It’s from what you said, sometimes it’s our own error, it’s our own intuition, maybe it’s a hidden bias, right?

Debbie Berebich: Yep.

Mike Delgado: Where do the data scientists begin? Maybe they’re beginning in statistics, things that they can watch out for as they’re beginning to work in the math?

Debbie Berebich: Yeah, that’s a great question. Let me give you other examples that will explain what other kinds of mistakes can occur. One of the most important things is how you select your sample. It could be a biological sample. It could be a population that you’re going to pull and see the statistics for you know who you think is going to win for President, etc. And so there are different examples I can give you where things can go wrong.

We saw a small size of a sample was bad for propagating the earthquake myth of standing underneath the doorway. There could be for example a sample bias ,it’s called the survivorship biased, which means when you want to analyze a bunch of mutual funds and see which ones are successful and what are the reasons why they are successful. But you forget to include which ones failed in that time. So, you’re going to end up analyzing something that’s wrong because your conclusions were related to only the ones that survived, and not the ones that failed. That’s one example.

Another one is spurious correlations like I said when things are correlated but they don’t cause each other. Another is accepting results uncritically. When your code gives you a number, don’t just take it for granted. Talk back to the statistics. Does it make sense? What’s your known hypothesis? And make sure you have a hypothesis before you analyze data, because often people just start, what’s called mining the data. And there’s such a thing as confirmation bias, when people find the answer that they’re already convinced exists, so they find the answer they’re looking for. So, you cannot test your model with only that data set. You must test it out of sample, or with a different sample so that you can test if your theory is valid or not. There was a statistician that said a wonderful phrase.  George Box, who said “All models are wrong, but some are useful.” I love that.

You reminded me when you said the political polls. I remember in February of 2016 after the caucuses in Nevada, Trump said that “Hey 46% of Hispanics are voting for me.” 46% of Hispanics, which is very surprising because you know just being Mexican myself, I knew he was not being very friendly to Hispanics. What happened, is his team manipulated the statistics to look like that. There were 1500 or so people that were polled, in Nevada. Most of them were caucuses, most of them were Republican, out of which there were 126 Hispanics that were Republican in that group. And out of those 126, 58 voted for Trump. So, then he said, you see 58 out of 126 is 46%. He made it sound like it was a national thing, and it wasn’t this completely biased sample. So, the way you create these samples is super important. The final two things that are super important are make sure your sample is large enough, representative of the population you want to poll, or the insight that you want to gain. Make sure it’s randomized, you’re not skewed towards say Republicans or Democrats or whatever you’re doing. Make sure you have a well formulated hypothesis before you start. And you can find unexpected and unreasonable conclusions that for a moment seem logical, so make you ask a colleague of yours to check on your conclusions, and you’re not committing confirmation bias. Checking all those things is incredibly important.

Mike Delgado: And I love your suggestion about checking with a colleague just to verify because we may not even know we have a bias, right? We could just be looking at the data, and going “oh, this proves what I was trying to get at.”

Debbie Berebich: Yeah.

Mike Delgado: But there was a bias there and without having someone else looking over their shoulder, looking over the model, looking at the work.

Debbie Berebich: Absolutely. I have a girlfriend, Cathy O’Neil, who published a great book called Weapons of Math Destruction. And it’s all about how we are responsible for creating the algorithms that we create for creating the samples and the statistics so all algorithms are inherently biased. For example, if you know the algorithm that predicted when should an airbag open in a car when an accident happens, was biased because all the equations and everything was tested only with men. And what’s suddenly happening is that when women were, and you know children on the passenger seat, in those cars, there were accidents and there was a risk of suffocation until they included women in the team and they designed under those conditions. So again, make sure the sample you are feeding to the algorithm that you’re working with, is representing one of the populations that you’re going to be working with.

Ethics takes a very central role in AI and machine learning because the way you prepare those algorithms to keep on learning and working is very much dependent on how you, your ethics and how unbiased your sample is.

Mike Delgado: I love the way that you talked about that and thinking about ethics and statistics, that’s something you don’t necessarily think about when I think about math. I don’t think about ethics, but when you’re talking about statistics and making claims, especially in the medical field, or like you just said automobile accidents, yes, ethics are involved in every aspect.

Debbie Berebich: Yeah.

Mike Delgado: Cause like you said, models can be biased. Like the ones you’re choosing could be biased.

Debbie Berebich: I think is sad and dangerous in the medical field. So, it must do with the anti-vaccine movement. Andrew Wakefield was the scientist that came up with it. I think it was in the 1980’s or forgive me for not remembering, but it wasn’t too long ago. And basically, he was a physician who claimed that the MMR vaccine causes autism. And lots of people believed in it. A lot of damage was done, sadly because that myth propagated and now the anti-vaccine movement continues to grow. And if you look at what happened, they took his medical license away, there was fraud because he was paid by lawyers and wanted to sue the vaccine companies.

The board of ethics and hospital did not approve the methods; all kinds of things. But if people had only looked at statistics of his study and you’re all welcome to just Google his paper, Andrew Wakefield, his study was only with 12 children, even though millions of children had received the vaccine with no problem. And out of those, 8 were the ones that he claimed had conceived autism from that. So only by looking at this small sample people should have discarded this paper, but sometimes the misreading of statistics can be very harmful for society. Ethics matters a big deal.

Mike Delgado: Thank you so much for that. We only have a short period of time left, before we go could you share a little bit about Metis and the courses that are offered.

Debbie Berebich: Of course.  Metis is at and by the way, I invite all your viewers and everyone out there, if you want to learn more about data science, we have an amazing free online live conference this Wednesday, 27th of September. From 10 AM to 10 PM Eastern time. We have amazing speakers coming to speak. It’s called Demystifying Data Science. All kinds of topics of interest to anyone, very intro and other more sophisticated. But you can choose and all the information is in the link that Mike is going to show.

Metis is a company that does data science training, so we have boot camps, consecutive boot camps, 12 week immersive that teaches you everything to do with data science to get a job in data science in the future. And the boot camps are in Seattle, San Francisco, Chicago and New York. We also have corporate training, where we train people in corporations in an aspect of data science and we have professional development courses, for those of you who are not sure you want to take a full boot camp, and want to take just a basic intro data science course, or if you want to go deeper in something such as deep learning or big data tools and what not, you can also take these professional development courses.

Mike Delgado: if you’re watching here on YouTube, check out the about section I’ll have the link there. If you’re watching on Facebook look in the comments for Check out all the different online classes that are available. I think it’s going to be the future as more businesses are going to need to equip their employees with backgrounds in data science, machine learning, artificial intelligence, so they can do their work more efficiently. And to be able to work more efficiently as AI and machine learning becomes part of their work.

Debbie Berebich: Absolutely.  Couldn’t agree more.

Mike Delgado: Debbie, I want to thank you so much for your time today. For sharing about ThisisMetis, as well as sharing some comments of statistical errs that are made, not only by normal people, but also statisticians. Thank you so much for sharing that.

I want to let all the viewers know that if you’d like to find out about upcoming on Experian Datatalk, you can also go to to learn about, more about Debbie and the work that she’s doing over at ThisisMetis. Debbie, again, thank you so much for your time …

Debbie Berebich: Thank you. And if anybody has any questions, just tweet at me I’m @DebbieBere, D E B B I E B E R E. Thank you again. It’s been a pleasure Mike.

Mike Delgado: And I’ll make sure to add on her Twitter link and LinkedIn links in the about section of the videos.

Debbie Berebich: Thank you. Have a great rest of your day.

Mike Delgado: Take care. Have a great next week.


About Guest

debbie-berebichezDeborah Berebichez is the Chief Data Scientist at Metis, physicist, and TV host for Discovery Channel’s Outrageous Acts of Science.

She has expertise in scientific research and advanced analysis and she has helped automate decision-making and uncover patterns in large amounts of data. Her passion lies in merging critical thinking skills with practical coding skills. She specializes in drawing connections between the approaches used in data science and the challenges organizations face.

Deborah has a Ph.D. in physics from Stanford and completed two postdoctoral fellowships at Columbia University’s Applied Math and Physics Department and at NYU’s Courant Institute for Mathematical Sciences. She is a frequent mentor of young women in STEM. Her work in science education and outreach has been recognized by the Discovery Channel, WSJ, Oprah, Dr. Oz, TED, DLD, WIRED, Ciudad de las Ideas and others.

She is a John C. Whitehead Fellow at the Foreign Policy Association, a winner of the Society of SHPE’s STAR Award and a recipient of the Top Latina Tech Blogger award from the Association of Latinos in Social Media (LATISM). She is also the first Mexican woman to graduate with a PhD in Physics from Stanford and engages in a variety of activities as a science, technology, engineering and math (STEM) ambassador.

Make sure to follow our Deborah on TwitterLinkedIn, and check out her website: Science with Debbie

Check out our upcoming live video big data discussions.