Wendy Martinez has been serving as the Director of the Mathematical Statistics Research Center at the Bureau of Labor Statistics (BLS) for six years. Prior to this, she served in several research positions throughout the Department of Defense. She held the position of Science and Technology Program Officer at the Office of Naval Research, where she established a research portfolio comprised of academia and industry performers developing data science products for the future Navy and Marine Corps. Wendy is also proud and grateful to have been elected as the 2020 ASA President.
+ Full Transcript
John Bailer: Welcome to Stats and Short Stories today. I’m Chair of the Department of Statistics at Miami University, and I am delighted to be here today with Wendy Martinez, incoming President of the American Statistical Association and also the Director of the Math Statistics Research Center at the Bureau of Labor Stat. I also have in the studio with me, my partners in crime, Rosemary Pennington and Richard Campbell from the Department of Media, Journalism and Film also at Miami University. Wendy, welcome.
Wendy Martinez: Thank you. Bailer: Well Wendy you’ve been doing this computational stuff in statistics for a good long time. Some of the earlier work that I’d seen that you had been involved with MATLAB, and one thing that I’ve noticed that over the course of my career- the tools that I’ve used have changed and some of the kinds of problems that I’ve been encountering when I think of as data- has changed. So, I’d just like to ask you to start this short conversation with what have you changed in terms of how you use tools and data analysis, and what kind of fun problems are you able to work on now?
Martinez: Well let me start with that one, this is sort of a personal story, I guess. So, I started working, you know, going to school later. I had a non-standard educational career because I got married and had kids first, and was in the army, and started taking classes here and there. At that time, believe it or not, I was using a Commodore 64.
Bailer: Okay yeah.
Martinez: Okay good. I was hoping you would say that you knew what I was talking about.
Richard Campbell: I do too. I remember.
Martinez: Yes. And so, believe it or not the storage was on a cassette tape.
Bailer: Oh yeah.
Martinez: Yes. So anyway, that was my first introduction to programming because we would get this magazine that had code for games that people wrote, and then we would type it in, and of course we’d have to debug it, and that was where I first got interested in that. So, then I started going to school full time, and got my bachelor’s degree, and master’s degree, and then started working on my doctorate in computational sciences and informatics with an emphasis on computational statistics.
Bailer: So, you were ahead of the curve Wendy. I mean you were doing data science when it wasn’t called that.
Martinez: I agree completely with you John. It really was what we call data science today. And the idea was that if you have problems that can only be solved using computers, how do you do that? So, I got my Ph.D. and worked on problems supporting the marine corps. And then my husband started working on his in the area of – he was also getting this computational statistic, but his focus was on how we analyze unstructured text.
Rosemary Pennington: So, what do you mean by unstructured text?
Martinez: Well that is just- so if you have sometimes- so say you have a medical record. Somebody might type in your temperature, I guess that would be a number, that wouldn’t be a text. Some name of a disease or something, but then you would have the doctor’s note, which would be just an information about the why you’re there and that would just be like a dialogue- a narrative, I guess. So, I work at the Bureau of Labor Statistics, and we have interviewers that go out and they have another field or a notes field where they can type in information about the situation or the call that they are on. So, it’s things like those, so there’s no structure to it.
Bailer: So, what kind of analyses were you trying to do with it?
Martinez: He was trying to develop ways to represent the text or encode the text so that you could compute with it. And so that’s what he developed. He didn’t do this, but one of the common ways is the back of words, or the term document matrix. We just represent every document as sort of the word frequency performed- the frequency of the words in the document. So, what’s interesting about that is it started- when I realized that this is such a rich source of knowledge and information that we don’t really take advantage of. And so, it’s just exciting because statisticians typically think of data as being numbers, but this was something different.
Bailer: And it’s gotten easier over time, huh?
Martinez: Yes. And I think more and more statisticians are getting involved with this as we see by the tidy text package you mentioned earlier.
Bailer: Thanks. That’s a shout-out to former Stats and Stories guest, Julia Silge.
Martinez: Thank you. And often times when you hear about big data, the example that’s used with big data is this text-based data, or unstructured text. So, I just think it’s exciting. And it’s not easy to work with because- I mean, I don’t want to say it’s not easy to work with. But how do you – if we use words in order to convey meaning, that’s really hard to handle in a quantitative way.
Bailer: Richard looks like with a great sense of relief to hear you say that.
Martinez: I thought he might like that.
Campbell: That’s right I do.
Bailer: Well Wendy that’s all the time we have for this episode of Stats and Short Stories. Thanks, so much for being here.
Martinez: You’re very welcome. Thank you for having me.
Bailer: Sure. Stats and Stories is a partnership between Miami University’s Departments of Statistics, and Media, Journalism and Film, and the American Statistical Association. You can follow us on Twitter, Apple podcasts, or other places where you can find podcasts. If you’d like to share your thoughts on our program send your comments to email@example.com or check us out at statsandstories.net. Be sure to listen for future episodes of Stats and Stories, where we discuss the statistics behind the stories, and the stories behind the statistics.