What is Biocomplexity | Stats + Stories Episode 339 / by Stats Stories

Stephanie Shipp is a research professor at the Biocomplexity Institute, University of Virginia. She co-founded and led the Social and Decision Analytics Division in 2013, starting at Virginia Tech and moving to the University of Virginia in 2018. Dr. Shipp’s work spans topics related to using all data to advance policy, the science of data science, community analytics, and innovation. She leads and engages in local, state, and federal projects to assess data quality and the ethical use of new and traditional data sources. She is leading the development of the Curated Data Enterprise (CDE) that aligns with the Census Bureau’s modernization and transformation and their Statistical Products First approach. She is a member of the American Statistical Association’s Committee on Professional Ethics, Symposium on Data Science and Statistics (SDSS) Committee, and the Professional Issues and Visibility Council. She is an elected member of the International Statistical Institute, an American Association for the Advancement of Science Fellow, and an American Statistical Association (ASA) Fellow. She received the ASA Founder’s award in 2022.


Episode Description

One thing that we always value at Stat+Stories is the story of, “How did we get here?”. Today’s episode follows our colleague, from work that she did in the federal government to now leading the charge at a biocomplexity institute. That's the focus of this episode of Stats and Short Stories with guest Stephanie Shipp.

+Full Transcript

John Bailer
One thing that we always value at Stats and Stories is the story of how did we get here? Today's episode follows a colleague from work that she did in the federal government to now, leading the charge at a biocomplexity Institute. That's the focus of this episode of Stats and Short Stories. 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. Rosemary Pennington is away today. Our guest today is Dr. Stephanie Shipp. Shipp is a research professor at the biocomplexity Institute University of Virginia and a member of the American Statistical Association's Committee on Professional Ethics, Symposium on Data Science and Statistics Committee and the Professional Issues and Visibility Council. Stephanie, thank you so much for joining us again.

Stephaine Shipp
And thank you, John, for inviting me.

John Bailer
You know, I really love seeing kind of the rich background that you had in coming into this new role, or your current role in the biocomplexity Institute, you know, ranging from work at BLS, could you tell us about your service? How did you get from there? Early on, what attracted you to kind of thinking about dealing with the important questions of data analysis as applied to economic and social problems? Maybe that's a place to start.

Stephaine Shipp
So I always like to tell the story that when I started at Trinity University in Washington, DC, I was a day hop, and that was way back, so that even that word dates me quite a bit. And I was a day hop the first year, because my parents decided I should just commute the first year of school. And I really thought my life was going to be political science. Maybe politics, maybe become a lawyer. You know, a very traditional route. I had this all planned out, but because I was commuting and I couldn't have the car on Thursday, I couldn't take political science, and so I had to take economics, and I fell in love with economics. And Sister Martha Julie was an amazing economics teacher. So, you know, serendipity, right? I would like to say my whole career has probably been serendipitous. My husband disagrees. He says it's also a lot of hard work, but I still and so Trinity, the Econ department, had a connection with the Bureau of Labor Statistics, and they would hire many of us out of Trinity. And so my first job was with the Bureau of Labor Statistics, and I was assigned to work in it in the office of unemployment and statistics, and so it's really interesting. And I started a summer job there. Happened to meet my husband. That's another story. Stayed there for a few years, but my heart had really been, having been an econ major, I just had this idea I wanted to work at the Federal Reserve Board. So I applied there, and almost didn't get the job, because when I called to ask that, I said I hadn't heard from you, they said, Oh well, your advisor said you're going to law school. And I said, Oh well, my plans have changed. I've got married, and now I think I want to work at the Federal Reserve Board. And they said, Okay, you're hired. So I started my job there, but it quickly became apparent that you weren't going to progress there if you didn't have a PhD. So I had a bachelor's degree. I was meanwhile in school at George Washington to get my master's degree, which I did complete, but had worked so hard to do that at that point I wasn't ready to. Suva PhD, so I went back to BLS and worked on the Producer Price Index Program to look at the modernization and innovation of that, which was really interesting work. And I developed a lot of good data skills as well, not only at Phillies work, but there and then, I think, through serendipity again, I had been out on maternity leave and then was invited to join the Consumer Expenditure Survey. And that sort of survey just sort of opened a whole new world of research and data to me that I hadn't seen before. First of all, you're looking at multiple societal issues when you use this data, not just spending patterns of households, but how do poor people, you know What defines poverty? How might you use expenditures? Why might expenditures be a better measure of well being than income? Just lots and lots of questions. And somewhere along the line, I had worked on that program for about 14 years. About halfway through, I decided to go back to school. My daughters at that point were in grade school. My husband likes to say that he was a single parent during that period. That's not entirely true. You know, they were swimmers and very active, and did a lot. So I did. I spent a lot of my time at swim meets, practicing or reviewing or reading economics, meanwhile, timing in between. And then once I received my well, actually, before I formally received my PhD, I received an offer to get moved to the Census Bureau to work on poverty survey program dynamics, which was a 10 year Longitudinal Survey along with the survey of incumbent program participation. I actually thought I would retire from the Census Bureau, but three years later, I was selected to be part of the SES career development program that then took me to the National Institutes of Standards and Technology, brought all my survey and data experience there, but introduced me to this wide, wonderful world of innovation. And the program that I worked on was the Advanced Technology Program, a very small program. It was the civilian DARPA at the time, but it was a program that President Clinton loved and showcased, which made us an easy mark for Republicans who had never liked the program since it began in 1989. So I started there in 2001. By about 2008, the program was being transformed and finally shut down. But meanwhile, I had moved on to the Science and Technology Policy Institute, where I met Sally Keller, who was then the director of the Institute, and loved the work there. We worked and did evaluations and all sorts of innovation studies across many, many different agencies. So while my career had focused on the Reserve Board, BLS, census, and then NIST, a lot of agencies, for many in the federal career, but all of a sudden we're working with NIH, and I'm working with different parts of commerce, and we're working with NSF. I just love that. I forgot to add that when I moved from BLS to census, I also loved that, because at the BLS, since you are a statistician or an economist or maybe an IT. I mean, that's it. These are your three disciplines, your three silos. When I went to the Census Bureau, all of a sudden there were geographers and sociologists and political scientists. I thought that was cool. So clearly, I love this transdisciplinary world, where there's lots of different ways of looking at the world. And of course, it's moving to NIST as well. But at Stippy, I was there for four or five years, and then Sally Keller, who's now the chief scientist at the Census Bureau, she had an opportunity to start the Social Decision Analytics Division at the Biocomplexity Institute at Virginia Tech. So we co founded this lab together, developed it, stood it up and discovered that our sweet spot for our work was really working with local government officials to use their data. And while the big data revolution had hit New York City and San Francisco and all the large cities, it was the smaller areas that were really hurting for data science, and so we started that. Five years later, our whole institute picked up and moved to University of Virginia, and that's where the institute is now. Our lab has always been located in the Washington, DC area, while at that time the others were located in Blacksburg, or now Charlottesville. But you asked what biocomplexity is, and it is the study of massively interacting systems, starting from molecules up through societies and populations. So ours, we obviously, were on the society and population side of things. So that was what I don't know, a five minute version of my career.

John Bailer
That is such a rich, what a rich set of experiences. I mean, the consumer expenditure data had to be fascinating in terms of just tracking what, what happens in society. And then you move to other contexts where you're looking at other features. I was really taken with, you know, the description of the division that you're in. Now, you know the headline is democratizing data to serve the public good. And you know, we had a guest on Stats and Stories many years ago who talked about the importance of official statistics as being a human right. I mean, just the idea that data informs, and then as your efforts, I see that community well-being, national well-being and official statistics are kind of these focal areas. Could you just kind of drill down a little bit on one of the community well being efforts that you've you've all been involved with?

Stephaine Shipp
We've been involved with a lot of different issues. We started first with the fire chief in Arlington. And I always like to tell this story, because, at the end of the day, it seems like, Oh, is that obvious or not? But he gave us full access. First of all, we got to meet with all the firefighters. We got to go to their command room. He even offered to have us go out on the fire trucks, just to really experience, you know, okay, this is how these data are born. These are how the data are lived. But when you look at 911, data, all of a sudden, it was a perfect initial data science project, because they have multiple silos of data. And he asked really good questions, but our very first step was, how do we link these data through time, space and well, geography, that is time and space. How do you link them together? And so we were lucky that we had a great and very creative group of students who were also very energized by this data, and we came up with a system for linking the data. Once you come up with that system, of course, it's easier to do that the next time, but the first time is always harder. But one of the questions the firefighter or the fire chief asked, very good questions, but he had this one. He says, I don’t know the daytime population of Arlington, Virginia, but I know the nighttime population. I know the residential population of Arlington, Virginia, but I don't know what the daytime population is, and we happened to map these data over. We took three years of data, mapped it by time of day, day of week, and when you looked at this map, you could clearly see the flows through Arlington County. Because what would happen like on Monday morning, you would start to see the number of 911 incidents surface in this particular area. And then you'd look at Tuesday, and it was at a different time of day. And as the week progressed, you saw that these 911 incidents were occurring. And you could see these were sort of representing flows of people through Arlington. And then the weekend, of course, was very different as well. So this one movie sort of really sums up our initial work on showing us the power of data, telling you more than just what those data were designed for. And so that's where we first began to look about repurposing data, and how those repurposed data really can bring to life a story about an area or community in a way that is sometimes surprising or unexpected. And of course, as those patterns begin to build and we gain more experience, we learn, we begin to learn at that but I have lots of other examples too.

John Bailer
So what are you, what's kind of the most exciting thing that you're working on right now in the biocomplexity Institute?

Stephaine Shipp
For me, we just finished up a study that, at times, sounded like fairly traditional studies, but Fairfax County, which is our neighboring county, asked us to look at the status of women and girls in Fairfax County, and that was a study that brought together both qualitative analysis and quantitative analysis. And again, we always have what we call a community learning through a data driven discovery process. And we alluded to that when we were talking earlier. This is a process in which the community not only asks the questions, they participate in the research, and that's always been our model. So we are not researchers that go into the community and say, we have these researchers, we have these research questions we want to answer. Just hand over the data, trust us. We're going to do the study. They are involved at every step of the way, and we're always sharing results back with them as well along the way, because a lot of times they provide insights in a way that we don't understand. We'll look at data and say, This doesn't make sense to us, and they'll be like, right away, oh, you know, the population shifted there. It's now a younger population rather than an older population. So that process has always worked. What also often takes the longest part of the process is identifying that problem. So you could say, okay, that your question is, what is the status of women and girls in Fairfax County? But how do you answer that question? And so you really have to break it out into sub questions and look at that. It was impactful, because the second part of the study was where we had community conversations with women across all. Real sort of levels of income and professionalization. That's where the data became alive. And I can't advocate more strongly that with data science having that qualitative piece is an important part of it as well. But what we found in this study is that the status of women and girls is, you know, there's still a long way to go, even though this has been an issue, at least going Fairfax County recognized it as an issue in 1971 the UN recognized it as an issue in 1979 and it's been, and actually, a recent World Bank report has shown there's been actually a slide back in part, perhaps maybe because of covid or because of others. So the Board of Supervisors in Fairfax County has promised to take our data results and actually begin to implement policy around those results.

John Bailer
That's pretty exciting to see research coming to practice like that. Right?

Stephaine Shipp
It is, it was. And we had an event on Friday, which was to celebrate Women's History Month, and there was a panel of women who also talked about their experiences. And those were awesome as well.

John Bailer
Wonderful. Well, I'm afraid that's all the time we have for this Stats and Short Stories episode. Thank you so much, Stephanie.

Stephaine Shipp
All right, and I'd be happy in the future to talk about machine learning or maybe more data science oriented examples, but that was a recent one.

John Bailer
Oh boy. No, see, there's like a trailer for a potential upcoming movie. So thanks. 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.