Walter W. Piegorsch is the Director of Statistical Research & Education at the University of Arizona’s BIO5 Institute. He is also a Professor of Mathematics, a Professor of Public Health, a Member and former Chair of the University’s Graduate Interdisciplinary Program (GIDP) in Statistics. Dr. Piegorsch’s research focuses on data science and informatics for environmental hazards and risk assessment.
Episode Description
The best thing about being a statistician,” he said, “is that you get to play in everyone's backyard.” That famous quote by John Tukey is optimized by our guest and the focus of this episode of Stats and Short Stories with guest Walter Piegorsch.
+Full Transcript
John Bailer: The best thing about being a statistician,” he said, “is that you get to play in everyone's backyard.” That famous quote by John Tukey is optimized by our guest and the focus of this episode of Stats and Short Stories, Where we explore the statistics behind the stories and the stories behind the statistics. I'm John Bailer. Stats and Stories is a production of Miami University’s Departments of Statistics and Media, Journalism and Film as well as the American Statistical Association. Joining me are panelists Richard Campbell, professor emeritus of Media, Journalism and Film and Rosemary Pennington, a professor in the department. Our guest today is Walter Piegorsch. Piegorsch is the Director of Statistical Research and Education at the University of Arizona’s BIO-5 Institute. He joined us last week and it's certainly a pleasure to have you here with us today. Well, I mean, you know, certainly, I've had a lot of great time talking to you in the past and I'm certainly having a great time. Now, the one thing that I wanted to talk to you about is you know, indeed, I wanted to talk to you a little bit about your experience in working in an interdisciplinary group. How about gearing up for working within in an interdisciplinary context, you've worked with toxicologists you work with biologists you work with geneticists, you know, heck, even stoop so low as to work with me, but you have this extensive experience and I'm also someone who values, collaborative work and that, and working with, particularly across disciplinary boundaries. So can you talk a little bit about how you gear up to contribute to projects where you have these, these interdisciplinary partners?
Walter Piegorsch: By the way John you weren't the worst
Unknown 30:34 collaborative ever had.
Richard Campbell: John, what's the best compliment you've ever had.
Bailer: My head is just swelling now I'm positive here.
Piegorsch: Yeah, and not just toxicologists and cancer people but yes, and, and but also geographers and and social scientists so it's my perspective on this is that if you're properly trained statistician, you are automatically trained to translate your skill sets across those interdisciplinary boundaries. I had a professor once and he was referring to what we call survival analysis, but he was basically saying listen, once you know what the data are Arad is the same as a light bulb, or I guess these days a transgenic mouse is the same as an integrated circuit. And that's the great point, is that if the data are providing or exhibiting a certain kind of features that is properly trained statisticians can look at those features and decide what kinds of techniques, technologies, methodologies, we have to apply to them, and they can work in many cases and my works been an example of this, they can work just as well on cancer biology as they can, geographic hazard assessment. So how do you gear up for that, you get trained properly I've been kind of pushing that for, for a little while here now. Get your instructors get your statisticians, get your your your teachers to have you recognize the statistics is one of the most translational disciplines, available, and think about not just applying it to x, but apply it to anytime you see something that looks like x,
Bailer: I would say that you probably do more than that though in terms of your you ended up learning a lot about genetics, and I bet you learned a lot about geography. And, you know, as you've talked about things like vulnerability in other contexts on another episode of our show. You demonstrated that you, that part of this collaboration was bringing not only kind of your strong foundational skills, quantitative skills, statistical skills but also you're, you're, you're diving in and trying to kind of get some, some degree of competency and mastery. At within your with Collaborate, your collaborators,
Piegorsch: Yeah you gotta want to have fun with your collaborators right
Bailer: On this podcast!
Piegorsch: Oh yeah, you did this, you did that but if you are having fun, then you're learning something, I learned something from the geographers I worked with, I learned something from the geneticist and microbiologist I worked with, and they were, I'd like to think wise enough to then start listening to what I was saying and learn something from me, and it gets back and forth and back and forth but I call it the upward spiral where you learn something, You figure out that that's new and that's different, it's nothing you've learned before, but if you're ready to be listening to it, and you're ready to collaborate and CO and just talk back and forth, both of you can be better scientists or better scholars, then you were a day ago and just keeps going, if you let it happen and just keep going up and up and up and up. And that's that's for me that's a fun part of what we do in what I do in mind, interdisciplinary work,
Campbell: Listening to both, you know, John, and we'll talk here. I've had a long career in academia and we've often talked about silos departments as silos, and one of the things that always struck me about John was how much he was involved with, with folks of Miami from lots of different disciplines. Is it better now, are those silos breaking down or is statistics just one of those areas that's naturally interdisciplinary.
Piegorsch: Well I think the latter is certainly true. If you really understand statistics you understand that it's got a certain mathematical and theoretical foundation and we need that. But it's also got a computational foundation as we get more and more statisticians being trained in just basic not invasive but advanced methods of computer science, and that's leading to how data science is expanding as its own kind of field with statistics is as a component, whether or not silos are breaking down, I know I've been in a few different academic institutions. Sometimes yes, sometimes no. I certainly understand that the know often occurs when funding when resources are stretched very thin. And then it's natural to kind of pull yourself back in. But here at the University of Arizona, we often say that we've got very low, low walls, kind of step over the wall that's, that's defining the silo, if you will, and then you can pop in and start talking to somebody and you can do this exact kind of upward spiral collaboration, that, that I was mentioning earlier, so my hope is for the future that you know, resources get tight, it's kind of tough but when resources open up, we'll see more and more of this, this expansion of collaborative science and collaborative scholarship. You know,
Bailer: One of the things I hear in your responses the idea that there's a technology transfer component that someone with this sort of foundational skills can, can, can, kind of, you know, reach across and apply it to their but, but the other feature of this and you've, you've illustrated this in some of the work that you've done is is this opportunity identification, where you kind of realize and recognize a problem where with your, you know with, there's this this nice synergy between kind of what you bring as the as the statistician with the content specialists, and finding this new problem and creating a new opportunity for development. And that seems like that's a pretty exciting, exciting aspect of this word.
Piegorsch: Oh, thank you. It's a very exciting aspect. I actually have actually used the term technology transfer a couple of times. Along these lines, so thanks John for just kind of reminding me of it as well. It's if you're open to it and you've entered in your training is solid enough and people you've recognized how broadly, you can actually apply statistical methods you can start telling really good stories, which is what you guys do, and on this very very successful podcast. You got to be open to that kind of thinking, and that's what we certainly try to do with our students at the UVA, but I would hope every stat statistics program everywhere is trying to give those students that kind of thinking,
I think I forget if I told you this or you told me this, John. But every data set tells a story. I think you told me this. And it's the recognition that that's true, that tells that gets you to look into it and see what's happening. And just don't accept it as a series of numbers accepted as something that's going on. And if you don't understand the science behind it. Find someone who does. And again that the upward spiral I was talking about. You're both going to be better, just better people that are scientists, but are scholars at the end of the day.
Bailer: Well that's a, that's a perfect way to bring this to a close, I'm afraid that's all the time we have for this episode of stats and short stories, Walter. Thank you so much for joining us today.
Piegorsch: Well you welcome John it's been an absolute pleasure. Indeed.
Bailer: 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 you can find podcasts. If you’d like to share your thoughts on the program send your email to statsandstories@miamioh.edu or check us out at statsandstories.net, and be sure to listen for future editions of Stats and Stories, where we discuss the statistics behind the stories and the stories behind the statistics.