Tom Braun is a Professor in the Department of Biostatistics and has been a faculty advisor for STATCOM for the past three years. Dr. Braun is an international expert in the design of Bayesian adaptive designs for oncology clinical trials, and he has more recently expanded his research into snSMART designs for clinical trials for rare diseases. Dr. Braun has collaborated with a variety of medical and public health investigators, including bone marrow transplantation, cancer of the mouth, breast, and lungs, periodontal disease, and the development of anthrax vaccines. Tom also is an active member of the University of Michigan committees working to address issues of incivility, rankism, and harassment in academia, and he also active in developing new pedagogy for teaching biostatistics and data science.
Emily Morris is a Ph.D. candidate in the Department of Biostatistics and former co-president of Statistics in the Community (STATCOM) at the University of Michigan. In addition to the leadership role, her involvement in STATCOM projects ranges from summarizing patterns in counseling visits at a local nonprofit to identifying optimal locations for mobile food pantries in Toledo. Her research primarily involves machine learning methods applied to brain imaging analysis.
Episode Description
Where are the best locations for food pantries? What are the patterns and use of a crisis call center? How can services be improved for the senior population of Wahtenaw County in Michigan? These questions share a common denominator, they represent data and analysis needs of community service organizations. Statistics in the service of the community is the focus of this episode of Stats and Stories with guests. Emily Morris and Tom Braun.
+Full Transcript
John Bailer
Where are the best locations for food pantries? What are patterns and use of crisis call centers? How can services be improved for senior populations of counties? These questions share a common denominator. They represent data and analysis needs of community service organizations. Statistics in the service of the community is the focus of this episode of stats and 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 Department of Statistics and media journalism and film, as well as the American Statistical Association. Joining me is panelist, Richard Campbell, Professor Emeritus of media journalism and film, Rosemary Pennington is away. Our guests today are Emily Morris, PhD candidate in the department of biostatistics. And Dr. Tom Braun, Professor in the Department of biostatistics at the University of Michigan. They're both involved in the STATCOM efforts at the University of Michigan. And they're here to share their experience with us, Emily, and Tom, thank you so much for joining us today. Thanks for having me here. Well, you know, you got to start with the obvious. So what the heck is stack comm
Emily Morris
So STATCOM is a student led volunteer organization, mostly in the department of biostatistics at the University of Michigan. We also have student volunteers from the statistics department from other public health departments. And essentially, we provide pro bono statistical consulting services to local nonprofits and government agencies.
Richard Campbell
So I think, Emily, you were a co author on on the piece that I read on benefiting the local community. So I'm an older person, and I may move down Arbor someday because my son and grandson lived there. And and I used to teach at Michigan. And so what's what what is the work you that you do, how's it going to improve my life if I decide as an older retired person to move to Ann Arbor?
Emily Morris
So I'm glad you bring up that that example of the project. It was one of the most, I think, directly impactful projects that STATCOM has participated in since I've been a student leader here. And essentially, it was surveying people in the community that were 60 Plus, to get an idea of what services would be useful to them, where could the in Arbor Area Community Foundation improve services they were already providing. And the reason that they wanted to do this work is because they had a very large grant, they wanted to allocate those funds in a way that was going to impact the people in the community, and not just the ways that they expected it to be helpful. So they were trying to directly survey the participants who would be impacted. And, you know, I don't know the status of that project and where exactly the funds ended up going. But I do know that they had a data driven approach to figure out where the services were needed the most.
John Bailer
So what gets people involved? I mean, what is so Emily, what what sort of attracted you first to, to start working with statcom.
Emily Morris
Um, so, to me, I've always wanted to I like to find volunteer outlets for things that I care about. And starting grad school, I thought, what better way to get hands on experience with with real world data, and also feel like I'm having an impact and be involved in my community. So that was the main reason I first volunteered for a project and then seeing the good work that was happening in some of these projects, and also, how appreciative some of the partners were was what made me continue, and then try to grow the organization and get more students involved.
John Bailer
So we're looking at sort of to two members of this community, that sort of different responsibility. Emily, you were one of the student leaders. Tom, you were one of the faculty advisors, what what were kind of your respective roles, I'll start
Tom Braun
The faculty. So there are three of us now. So stack up used to start out with basically one faculty advisor, and then it became two and now became three. So you can see that we're growing quickly. I see my role is simply to provide a path for the students, I try to step away as much as possible. So we have a monthly leadership meeting. Well, the faculty will chime in there. One faculty shows up the first meeting that we have with a client to just sort of help get the conversation going and, you know, answer questions that maybe students haven't had enough experience I'm asking quite yet. But then the group, the group comes together, they start meeting with the client analyzing the data producing a product. I stay out until they say we really need your help. Dr. Braun, I look at the report. I do feel like I want to give my stamp of approval on the product that goes to the client, but it's it's an easy job right now. We've got just great students that just do such great work.
Emily Morris
Yeah, so I'll say as a student volunteer, just one of the participants on a project Wish I was at the beginning of my time at STATCOM, that's the main role is to do the analysis. So you get a lot of experience with real data, working with people who typically are not statisticians, which might be different from what you're getting in coursework. And then as a student leader, the role is to do anything from you know, communicating with potential clients, determine if it's a good fit for STATCOM, or maybe a different organization on campus that does data analysis. And then it can be leading a team leading a project, making sure that team stays on track. So sometimes it's more administrative, having weekly meetings and you know, prioritizing different parts of the analysis. Sometimes it's doing the parts of the analysis that maybe are, require more expertise, or a little bit more of a literature search, things like that, especially as a more senior student. Sometimes we rely on other PhD candidates to help with the more intense statistically intensive projects. But as a student leader, yeah, it can be kind of a wide array of just running organization, as well as doing the projects.
Richard Campbell
One of the things that I've learned from john, is, came out of a phrase in one of your articles, I think it's a famous phrase about working in other people's backyards that the way that statisticians are involved in a lot of different fields. I think really, before I started working with john, I always thought interdisciplinarity was the domain of the humanities. And there was a lot of you know, I certainly study it across a lot of different disciplines. And in my graduate days, can you talk a little bit about the benefits of working on other people's backyards, both Tom and Emily, I'd like to hear your views on that.
Tom Braun
It's probably one of the best parts of our job. And the fact that so I've been doing this for 21 years, not just with statcom. But I, I started on bone marrow transplant, I've never wanted to go to medical school. But I feel like I know more about a bone marrow transplant outside of my collaborators. And that's just a unique benefit of just working with people and transplant. The flexibility in what we do is just, it's it's amazing the application of having a quantitative brain and helping others understand what's in their data. It's just, it comes with a lot of privilege. And I brought I'm glad you brought this up, because I think one of the hardest things to teach to students and we mostly graduate students, is just that privilege of being in someone else's backyard, and to not take it for granted. And the analogy I always use is that I'm the mechanic with somebody's car. And if I come and say, yo, you need a complete overhaul in your engine, and I can't explain why. I'm not going to be out of business. So I try to teach folks that if you're going to work in someone's backyard, you also have to tell them what you're what you're doing. So that they believe you, and they'll come back.
Emily Morris
Yeah, I'll just get a second that. I agree it's never boring, there's always something new and interesting to work on. Because you can collaborate with people on lots of different areas. And in terms of statcom to I think it's really interesting to work with different types of organizations and, you know, people who have statistical expertise and those that don't. And, you know, really engaging with different communities and figuring out how you can communicate what you're doing. So that's also a big part of I think learning through these.com projects is, I find, at least if I'm able to, you know, carefully articulate what I'm doing to someone who doesn't have any statistical formal education, then I know it a lot better. And I feel more confident taking those skills out after grad school.
John Bailer
Yeah, I love the one quote, I read it one of the pieces that that you all had written about helping to make the world a better place one analysis at a time, I you know, that there was a nice thing that resonated I thought that was attractive. Now statcom is is a, it appears at more than one academic home. This is a this is multi campus. Can you talk about kind of the breadth of where of different institutions where we're a stack comms organizations are found?
Emily Morris
Sure. So it was founded at Purdue in I think, the early 2000s. And I believe that they still have an active statcom chapter. I have touched with them a few years ago when they were doing a lot of K 12 outreach programs, a little bit different than what we do at Michigan. And I know that there have been a few other chapters, I believe at UNC
Tom Braun
Columbia's trying to start one up. Right. And that's actually I think, a goal of our statcom group is we sort of see ourselves now was the national impetus for statcom staying alive. It's listed on the ASE website somewhere, but if you go to that link, the website was last updated and you know, 2008 so the crown student leadership, I think is working really Hard to make more students that other universities want to do this and develop their own chapters. We've had one student who went to school here and then went, I think, to get their PhD somewhere else. So that that, you know, spreads the word. I like to say, I think we are we are the most active chapter, I think we believe that's true. I don't know that's true. But it's, we'd like to get changed that we'd like to make it a national organization again.
John Bailer
Well, having having 100 students involved, that sounds like a, that's a pretty big investment of kind of energy from your community.
Richard Campbell
So I'm, I'm kind of wondering there students involved in this. And there are, I'm imagining full professors who are our advisors. So in my time in Michigan, we were actually discouraged from getting involved in community projects while we were working our way toward tenure. And I'm wondering if the situation is different now, or?
Tom Braun
Yeah, this is always you know, the reason I came into statcom was because it was part of one of the service roles in the department, and they needed someone and I think I'm someone who expresses more of an interest in these sorts of service roles, and some of my colleagues, so I fit really nicely. So there are three of us who are leaders right now of effective leaders, two of us are full professors. One however, is not but I agree with you, I think there's only a certain amount of service that a junior faculty member can do. And there's other things they have to do to get promoted. And we want that to happen. So it's, it's an actually a really good point, because, you know, as faculty rotate on and off of service projects, I'm not going to be in SATCOM forever. And I would highly encourage it. Like I said, this is not a lot of work we have we have students who keep the organization going. And so I don't think it's going to cut into it hasn't I mean that? How do I judge this? for Tom Brown, 20 years ago? How much time this would take out of my research and things? It's just it's a, it's a great experience for an educator. I'm outside of the classroom.
John Bailer
You're listening to Stats and Stories. And today we're talking about statistics in the community with the University of Michigan's Tom Braun and Emily Morris, you know, so no, I'm not, I'm thinking, Emily, now, I got to do this, I have to do this in Miami, because I got, you know, so there, there are a number of schools around the country where you don't necessarily have PhD students that are available. You know, there are some schools that are strictly liberal arts, colleges, there are others that are going to be, you know, have a mix of graduate programs, maybe the highest degree might be masters to Can you can you imagine kind of other models, I mean, Michigan, you have a very large graduate student population in your department, as well as students kind of at all levels. Can you imagine an approximation to this, that that could be implemented at different different types of institutions?
Emily Morris
Yeah, absolutely. I think that there are a few key ingredients that help us come chapter really thrive. But not having graduate students is definitely not an issue. I think having, you know, a somewhat large base of students who are interested in helping the community and engaging in, you know, community service type projects like that is super beneficial. And certainly undergraduate students can partake in almost all the projects that we do. So I think having the faculty leadership, when you're first starting a program is super important to keep it keep those students on track and to give it some direction, but it'd be especially because of the student turnover. Having that leadership in the apartment is is helpful, and having the leadership in the organization is helpful. And then another thing that I think is really important for a second chapter to thrive is having a source of projects. And you're at the University of Michigan, we're lucky that we have a place called the Ginsberg center. And they focus primarily on fostering community University partnerships. And so that can be you know, common a lot of different forms. But a lot of the organizations look for data analysis health, and so they can kind of funnel those to STATCOM and they have a lot of people that are dedicated to working with community partners. And they also have a lot of the training and social work and things like that to create really productive partnerships. So I think having a lot of users that universities probably have a similar program to kind of find that community outreach, all those opportunities, but there also can be other ways to have the source of new projects. So I think if you have someone who's kind of creative or has a lot of connections in the community and knows that there's a need in the community for some, you know, data analysis help, that's that's definitely beneficial. So those two things, having students, a large enough body of students that will volunteer and having some sorts of projects really helps us come to after thrive.
John Bailer
Yeah, I think we know some undergraduates that had experience doing research and can get involved in that. So That can be quite successful. We've we've done some work at here with the Health District, Local Health District. And also we've done some work with the Butler County Coroner. And and one characteristic that's common is they have data, they have analysis they have needs, and they don't have resources to, to basically do that work. So I, you know, I think it's a if you don't have to look very far, but I think that the Those are some great advice, Emily, in terms of trying to, to build that as, as the pool of people that are interested in studying statistics and data science grows. I and you know, I know, Richard's been involved with this in terms of local journalism, even in trying to promote how students can be involved in that type of outreach. I was curious, I saw that you had mentioned one of the projects that you started, but one where you saw some impact, and you were really excited. And it was just a really fun, challenging problem.
Emily Morris
Sure, yeah. So one of the products that I think You teased earlier was trying to find an optimal location for a mobile food pantry. So that was a really interesting project for a lot of different reasons. It was statistically challenging, I learned a lot, I had not done any sort of spatial optimization before. So I found it to be, you know, interesting. From that perspective, it was also, you know, very interesting question to think about this very simple idea of, we have a mobile food pantry, we have a general geographical area to put it in. But where should we go? And what places Can we really serve the community the best, and so that that project was interesting, because, you know, most places in that position would just, they would put them up a food pantry in a location based off of convenience, where their community partners that are willing to let them park their mobile food pantry, where have they put it historically, things like that, but to say, you know, here are some characteristics of areas that we've been collecting for, I think, about 10 years at the time that they reach out to us. And then also usage data, things like that, and then try to quantify, you know, what locations work the best, and where should you be visiting more frequently, was very interesting. And I think there were some surprises that some locations, or analysis showed, they should be repeat visiting more often than they were and, and things like that. And it's also a very, there's a direct impact, and they can also reevaluate really easily. So if we suggest some locations, and they tried them for a year, and you know, they saw their usage go down, they could say, okay, maybe the metrics we looked at weren't really what we actually wanted to measure. And, you know, try to approach it from a different perspective, and try different locations. So that that one was really exciting to me. And I think, a very fruitful collaboration for both the students and for the local food pantry organization.
Richard Campbell
I want to find out what kind of connection you have with the journalism in your area, I think it's really important that these kinds of partnerships, and these kinds of outreach programs get publicized. So the general public knows about them. Do you have a formal relationship with journalists at all? Or is this something you rely on the Michigan PR department to get the word out about this? And and do they I mean, I think we live in a time when there's a lot of resistance to science and data and evidence. And I think it's really crucial. And we're also we've lost a lot of reporters in the last 15 years. So I think these are really important issues. And I wonder if that's if statcom and the Michigan version of it has as an outreach to journalism program, or you have journalists that are kind of willing to cover these not just the one off story, but to cover them over time? Because this takes time to do these right?
Tom Braun
The short answer is no, we actually don't. And as you're saying, this is sort of like the slap in the face. Like, why why haven't we talked about this? I think we're at a point now where we are expanding we statcom are expanding so quickly that we have talked about this. I mean, I'm an academic, when I want to get the message out, I think of journal article, we should read a journal article, but as I'm talking with you, I'm like, that's not our audience. There's a whole public audience out there that would benefit from something in the Michigan daily, for example, the student newspaper or Michigan records. So no, and as a jockey do like, well, maybe we should find a grad student in the journalism school school and and see what they can do with us. For us, we would never even thought about actually, it's, it's, I mean, my biggest interest now in my life, and my career is just trying to teach people how to use data and understand the statistics and the proportions that they read in the papers and the websites mostly nowadays. So it's it's a wonderful idea. And this is why we do these things because now I've got a great idea Emily and I do to go back to our leadership.
John Bailer
Yeah, I'll tell you I have I have just dearly loved collaborating with my My friends here in journalism, I mean, you know, Richard and I have had a, you know, the things that we learned when we start, we started working with each other, I think was surprised us both were surprised on two dimensions, one, how much we had in common and two, how much we thought about things differently. And, you know, in those, those are both really good things. And I, you know, Rosemary has also done a wonderful job of, of helping me think about the way we communicate what we do. So I would suggest that, you know, embedding a journalist with with your, with some of your teams might be a really interesting component of, of getting the word out. And it also is as a stat community, just thinking about how to tell stories in a way, a non journal article way, which is a really different kind of communication.
Tom Braun
Yeah, our dean's office for a while he's for the faculty was trying to have these storytelling sort of workshops where they had someone who was not a scientist, I don't think those exist anymore. But I agree, I think we're used to telling stories in a certain way that bore the heck out of our large part of the population. So a more general audience approach is crucial nowadays.
John Bailer
So I'm curious, what's been one of the hardest, what's what's one of the hardest projects you've worked on? The most challenging?
Tom Braun
So what I can tell you one of the hardest projects that we ended up not working on, I think, is just to show you, you know, how things have moved around in in statcom, is, we were approached by someone to look at the Ann Arbor police department was collecting data on traffic stops. And you can probably imagine where the story is going. And with all the racial unrest in this country, and the discussion around black men being stopped by police more often they said, Well, we have some data here. We don't know what it looks like, could statcom help us and we were very excited. And then as we thought about it more, we thought about the fact that this is such a highly polarized issue. We like to think that we're unbiased statisticians, but it was a really good discussion of, can we really be unbiased as analysis analysts of these data, if we have a certain social opinion right now. And we're also very worried about the students, we talked about publicizing results, these kind of results would probably have made the newspaper in Ann Arbor, I don't think graduate students are quite protected enough. Who knows what gets leaked in the press. So that was a really challenging one. For us. It's not so exciting, but we thought probably wasn't appropriate for our organization. The other difficult one that Emily and I it's, it's also wonderful, but it's been very challenging, is working with an organization looking at spaying and neutering of pets, we'll see right now dogs, which may not seem like a huge public health issue. But there is a real problem in dogs with bigger dogs. As they get older, they can't move around. There's an obesity problem in animals. And there is a belief that there's a connection with spaying and neutering occurring too early. We've been doing this project for over two years. Emily, I looked at an email earlier today. Usually our projects are six months ish. But this one's been going on for two years. And it's been fascinating to work with a group of individuals who aren't quantitative but know so much about their field. And the benefits on both sides, I think have been enormous. It's been great for the students to see that. You can't just show them a table of odds ratios. And they go, Oh, there's the answer. It's taking a lot of work to communicate the results.
John Bailer
Emily, would you put what would you what was what's been your take on the spay neuter story?
Emily Morris
Yeah, I mean, for me, it's the longest running project that I've worked on. And hopefully will soon be wrapping up to two years seems like plenty of time for graduate school experience for some. Yeah, I mean, it has been challenging. But I would say it's also very rewarding. And, for me, personally, rewarding in the learning how to communicate and work with a team that's pretty diverse in their backgrounds. The statistical analysis itself is, you know, interesting, but not not extremely novel. So it's more collaborating with the team and communicating this results. That has been what is what's, that's what makes it the most challenging, because I would I would agree, I think that's probably the most challenging project that I've worked on, in my time here. And sometimes projects can be challenging for different reasons, if the partners are really excited, really invested have a lot of questions, but maybe they don't have the data to back up those questions. That's a common problem. But in this case, we have really rich data. They're really excited about what we're working on. It's more trying to communicate clearly and have you know, a game plan that we agree on that they also understand and think is the best path forward.
John Bailer
Well, thank you so much. That's all the time we have for this episode of Stats and Stories Emily and Tom, thanks again for being here. 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.