Eric Vance is an associate professor of applied mathematics at the University of Colorado Boulder and the director of LISA (Laboratory for Interdisciplinary Statistical Analysis), where he trains statisticians and data scientists to move between theory and practice to collaborate with domain experts to apply statistics to transform evidence into action. He is the global director of the LISA 2020 Network, which is a network of 30+ statistics and data science collaboration laboratories in 10 countries in Africa, South Asia, and Brazil. He is an Elected Member of the International Statistical Institute and a Fellow of the American Statistical Association.
Olawale Awe is an Elected Member of the International Statistical Institute (ISI) and a Fellow of the African Scientific Institute, USA. He is an Affiliate member of the African Academy of Sciences (AAS) and an immediate past Council Member of the International Society for Business and Industrial Statistics (ISBIS) (2017-2021). He is the First LISA Fellow and presently the Global Vice-President of Engagement and Public Relations in the LISA 2020 Global Network of the University of Colorado, Boulder, USA. His research interests include Computational Statistics, Machine Learning, Time Series Econometrics and Statistics Education. He has served on some important ISI committees and has facilitated several capacity-building workshops and seminars globally. Olawale holds a PhD in Statistics from the University of Ibadan, Nigeria, and an MBA from Obafemi Awolowo University, Ile-Ife, Nigeria. He is the lead editor (with Kim Love and Eric Vance) of the soon-to-be-released book titled “Promoting Statistical Practice and Collaboration in Developing Countries” by Taylor and Francis Group.
Episode Description
In many countries in the Global South, partnerships and collaborations are crucial to moving forward projects of various kinds. A network based at the University of Colorado Boulder has facilitated the creation of statistics and data science collaboration labs in 10 countries, The LISA 2020 Global Network and it's efforts are the focus of this episode of Stats+Stories with guests Eric Vance and Olawale Awe.
+Full Transcript
Rosemary Pennington
In many countries in the Global South, partnerships and collaborations are crucial to moving forward projects of various kinds. A network based at the University of Colorado Boulder has facilitated the creation of statistics and data science collaboration labs in 10 countries. The LISA 2020 global network 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 Rosemary Pennington. 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 as always is regular panelist John Bailer, emeritus professor of statistics at Miami University. We have two guests joining the show today. Eric Vance is an associate professor of applied mathematics at the University of Colorado Boulder, and director of LISA, the laboratory for interdisciplinary statistical analysis, where he trained statisticians and data scientists to move between theory and practice. He's the Global Director of the LISA 2020 network, a network of more than 30 statistics and data science collaboration labs in 10 countries in Africa, South Asia and Brazil. He's an elected member of the International Statistical Institute and a fellow of the American Statistical Association. Olawale Awe is an elected member of the International Statistical Institute, a fellow of the African scientific Institute, and he's an affiliate member of the African Academy of Sciences. He's the first LISA fellow, and presently the global vice president of engagement and public relations in the LISA 2020 global network. His research interests include computational statistics, machine learning, and statistics education. The two are joining us today to talk about the LISA 2020 global net Network. Thank you both for joining us this afternoon. Great to be here.
John Bailer
It's a delight to have you both joining us. Thank you for taking the time. You know, during this time of year, I was saying we're going to have this in the World Cup of statistical collaboration, that's going to be the focus of our discussions today. And I promise you, we will not go to penalty kicks, nor will we need extra time we're going to be able to do this and the time we have a lot of you know, I would, I'd love for you to just just start with helping kind of defined orientation for us. And in particular, I think it would be interesting to hear your take on the idea of statistical collaboration versus Statistical Consulting.
Eric Vance
Sure. So I think what we do or what we aspire to is collaboration, where we are team members on a research project or on a project to answer a policy question or make a business decision. And so I think that distinction between consulting and collaboration, you know, everybody has their own definition. Just personally, I think that the two main differences are that in collaboration, you're a team member, whereas in consulting, maybe you're, you're an outsider. And then the second, and maybe more important for me, is the focus of the work. So I think when we are collaborating, our focus is on solving the research problem, or making a business or policy decision. Whereas in consulting, maybe work, our focus is more on the statistics, and the technical aspects.
John Bailer
So let me do a quick follow up to that. And maybe Olawale, you might be able to share an example of some of the collaborations that you've been involved with.
Olawale Awe
Yeah, thank you very much, John. I want us to start as a faculty member, I saw that a lot of people were just doing their own thing, things in their own ways. And then way back, people who were researchers in my university at that time, have been contacting me, consulting me. And at that time, it wasn't really a collaboration, like Eric said, just consulting just dumped their data sets at your table. And I found that this is not the right way things should be done. And it wasn't until I met Eric, and we, I traveled to the US and I began to understand what collaboration really means. So when I came back, when I traveled to the US to be trained by Eric and LISA, I came back and then I one of the first collaboration projects I worked on in Nigeria was a microbiologist that came wanted to know what said And after Ted, the contents, the quality of paints, you know, different kinds of paints qualities. And when he wanted to apply, she subjected the paints to different treatments like environmental factors. And she had on that experiment, she applied some chemicals for the paint. And then she tested over time to be able to know what really affects the quality of paints in these according to different environments. So she brought up projects, and I was able to help her analyze the data set, she became part of the team, and we became team members together. And then the project tried to test on the paints. But then she was very happy, I was able to produce three major publications from that very project. And we have been in touch. So now
Rosemary Pennington
I'm wondering how the LISA 2020 global network got started, what was the sort of impetus for this focus on these collaborative labs.
Eric Vance
So, the impetus is that I wanted to provide an international experience for my graduate students. So before I went to graduate school, I had the fortune of being able to travel around the world, I traveled around the world three times for five years, and visited 67 countries, and had an excellent time and experience traveling internationally. And so I wanted to provide opportunities for my own graduate students to travel abroad. I also wanted to really expand this idea of the statistical collaboration laboratory, where students got great experience working on projects, and then domain experts got better research results or better decisions, because they were able to collaborate with statisticians. And so the first thing I tried was to just have some exchanges between graduate students from different labs, you know, with our lab and in different countries. And I told the graduate students, if you want to go somewhere, find a lab, you know, find the exchange will make it work. They weren't able to find any labs outside the US and Canada that actually use graduate students to work on projects to collaborate on various projects. And so those two ideas were combined with a third idea of actually building statistics and data science capacity in developing countries, based on some international projects I've worked on, and recognition that the local statistician who understands the local context of a problem is like the best kind of statistician. And so all three of those combined, you know, in in 2012, brought me to the idea of creating a global network of statistical collaboration laboratories,
John Bailer
Congratulations on, kind of, this work that's been done, you know, we see that it's expanded to 10 countries and 30 plus, plus labs that have emerged from this effort. I was curious if you could talk about some of the impacts that the LISA has had, it's in some of these institutions. And maybe Olawale, you would like to kind of start with this is kind of the first person to have gone through kind of this initial impact. And then maybe we could follow with some other programs that you would like to highlight.
Olawale Awe
Yeah, thank you so much. LISA 2020 has really had a lot of impact on statistical education in developing countries. One of the impacts, I can say, is being able to move from theory to practice at the initial stage when I came on board, when I started as a faculty member who saw that a lot of people were doing, just purely care, no consulting, no collaboration, ethics. But when these trends came on board, we began to really know how to really do collaboration out to practice the other side of statistics. That is the practical aspect. And a lot of people are happy about that. And it has also improved collaboration with domain experts. Like I just mentioned in my last answer, collaboration we were able to publish in high impact Jana Dan, what individual I would have been able to do without the stats labs. So, in a lot of all these 35 Plus institutions, you can see that through the instrumentality of LISA, the research arm has been boosted, the research quality has been boosted. And we have more publications in high impact journals, because they are able to meet with the statisticians, professional statisticians in this laboratory spread across the world. And through that people begin to appreciate statistics more. And that's one other major impact I can keep on counting. Another one is providing critical research infrastructure to the institution's. Like I said before, people just contacted statisticians on their own, not in a professional way, or perhaps they do it themselves. But now they have a place to go, especially in developing countries. So they can have, they can be sure that they are walking will benefit from expertise of statisticians in the laboratories, then we're able to transform help also in further transforming evidence into action. And, of course, this actually directly contributes to development in all these countries, because the statisticians that have been trained in this lab mentored from Colorado are able to help others, it has ripple effects, you know, all like before, when they will just give some optimal results, but now they have mentors from the US. You know, sometimes when they cannot handle those projects, they can write, they can collaborate or commune with the community, the lessor community, and then are able to solve the problem together in any part of the wall. And then through that we are helping to solve policy problems, societal problems through statistics and data science. So many impacts are there including mentoring young students, many of the students are truly sad, especially in these developing countries, they've gotten jobs after graduation, and they testify that's worrying or for LISA, they wouldn't have been able to get those jobs. And so it's really, really a breakthrough, a major breakthrough in the developing world in terms of statistics, education.
Eric Vance
Yep. So one example of an impact is on the people on the statisticians who work in the labs, there was a student at the University of Ibadan, laboratory for interdisciplinary statistical analysis, UI, LISA, who, you know, worked on several projects, and then graduated. And he was one of more than 1000 applicants for a government statistics position in the Nigerian federal government. And there was one position with more than 1000 applicants. And he got the job, because of his experience, transforming data into action, making data into useful and actionable decisions. And that was the experience that he got at the UI, LISA lab that allowed him to be the one of more than 1000 applicants to get the job.
Rosemary Pennington
You're listening to Stats and Stories. And today, we're talking about the LISA 2020, global network, with Eric Vance and all the Olawale Awe. I wonder, you know, there's so much discussion in sort of development communities in various fields about the asymmetry sometimes appearing in these projects, right, where some experts from outside will come in and try to work to help a community. And it's not always, or has not always been sort of as community focused as it might be or informed. And I wonder, given the work that you've been doing, and the success that the LISA 2020 network has had, what have you learned from this experience about how to sort of Forge successful collaboration, when you are essentially, you know, in some ways coming in from outside to sort of work with people in these communities.
Eric Vance
So the LISA 2020 network really was founded on the idea that to do good statistical work, you have to understand the context of the problem. And so, if there is an American who's working on analyzing data from, let's say, Nigeria, the American Statistician might have the technical skills but lacks the context. And so from the very beginning, we decided that we wanted to identify the local talent in statistics, and then help amplify that to build the capacity of the locals. statistician's to collaborate on research projects and to create business and policy decisions. And so we don't have that issue where we have a foreigner coming into a local, local community and working on a project. Instead, it's a local statistician, working with local domain experts and stakeholders to have a local solution to whatever challenges they're facing. You know,
John Bailer
I really love the idea of statistical capacity building as being kind of a core mission. And many professional organizations have that as one of their strategic priorities, including isI that you both have, have mentioned as part of your, your connection to the statistical community. So the question is, alright, so how do you develop these statistical collaborators? What are some of the things that have to happen to be successful in growing the capacity to address these teamed collaborative projects?
Olawale Awe
Yeah, we ensure that these labs, you know, last of the lab, they operate on that three tenets allow one, they have regular shots causes. So statisticians in the lamps or in the local institutions have been trained locally, through the instrumentality of LISA, wherever LISA is, they're trained and ensure continuous regular short courses, seminars, short courses, educational workshops, to the overall academic community in the respective institutions. And then also, apart from that, there are some graduate students that have to go through LISA, before they graduate, you know, they go to LISA to learn how to collaborate, how to work on projects, and how to attend to domain experts, they learn both the technical and the non technical aspects of statistics and data science. And so these efforts to ensure that people are being continuously developed in order to become better by the time they graduate or be able to stand on their own. Eric, we want to add to that.
Eric Vance
And so I would say that, the first thing is to recognize that collaboration is important, and is a skill that can be learned. And so I think that's part of an attitude of collaboration. And so that's really the first thing is that we, we find the technically skilled statisticians, who then also have a passion for making statistics more useful in society and to use statistics to help development. And then we equip them with the model of a stat lab. So as all the while I mentioned, there are three missions of every LISA 2020 stat lab. The first is a focus on training students. So to use projects that come into the stat lab as opportunities to train students, and the students learn both more technical skills that they have to implement on projects. They also learned collaboration skills. The second is to serve as research infrastructure for their university for their community. So to provide the services of collaboration to work on teams to enable and accelerate research, and make business and policy decisions based on data. And then the third is to teach short courses and workshops to build up statistics capacity, you know, outside of the lab, widely in the university or throughout the community.
John Bailer
So, you know, I'm curious if, if you've seen any increase in the number of students studying statistics at institutions that have have started LISA lab,
Olawale Awe
We have seen a lot. That's part of the impacts of LISA, LISA 2020. In fact, a lot of students now apply to those institutions, just because there is a lab there. So that's, I think it's awesome. And to summarize, the things that are being done in those labs and target CCC courses causes collaboration and consulting, causes collaboration or consulting. Those are the three major panics that have three major activities that are being done in those laboratories. So a lot of students are interested, a lot of them want to go to the institutions where the laboratories are to, to enroll, so that they can gain this experience, because over time they will suffer from just learning theory without practice. And I remember when I came back, I had to organize a workshop just about 10 years ago, when I taught out nationally, and at that time, in lot, a good number of drivers, students did not know what r was, they will be asking the letter R or Z a software, or what is that I let it in the US myself through Eric and I came back and began to train a lot of people. And right now, a lot of students can really, really work with our and other stuff like Python, SQL, science, a lot of you.
Eric Vance
And I was inspired to go into statistics graduate school, so that I could work on lots of different types of projects, because statistics is applicable everywhere. And so, you know, with the stat labs, we have lots of projects coming in, you know, lots of different different applications of statistics. And, you know, just personally, that's what drives me about statistics is being able to solve different types of problems to help people make data driven decisions. And, you know, I don't think I'm alone, I think that also is attractive for many other students. They want to contribute to the world using their technical skills, and studying statistics. And working in the Stat Lab is a great way to do that.
John Bailer
Yeah, when I, you said earlier, the idea that collaboration is a skill that can be learned. So there's still hope for Rosemarie and I, to figure out how to work together? Is that what you're saying?
Rosemary Pennington
You know, your last call?
John Bailer
Well, that's yeah, maybe well known. You know, one thing that I, you were talking about this, there's a lot of history of teaching technical subjects, you know, teaching kind of formal analysis there, they're often these kinds of data practicum consulting classes that have existed. Uh, one thing that I think that I've seen in which stuff that you've written and work that you've done, has been really nurturing this aspect of collaboration. Some have called these skills, career success factors. In some contexts, others have called them soft skills. I think that's, that can be a pejorative, I don't, I don't like that, I really do think of them as what you need to really be impactful in in work so so you know, can you just just give us a sort of a short reflection on some of the ways that you help build some of these, these complementary skills to nurture collaboration to go along with the technical training people have had?
Eric Vance
Sure. So I've developed a framework called the asker framework, it has five components: attitude, structure, content, communication, and relationship. And so these are five aspects of collaboration. And they, within each of them, we can drill down and learn, you know, identify skills that are necessary, and then learn those skills. So, you know, just briefly for attitude, just the attitude that I am here as a statistician to help this domain expert, advance their research, or to publish a paper to make a discovery, or to make a data driven decision. And that I have, you know, lots to bring to the table, and I have lots to learn as well, I can learn more about technical skills. And I can learn more about the domain. So some, like those are some attitudes that are really helpful for collaboration. For structure, we've used Doug's Zaanse power process, and we call it the power structure, which helps us structure meetings, it's to prepare for the meeting, open up the meeting certain ways, do the work, and the meeting, and then reflect on how that will work. So that's power. And that's been really helpful to kind of give us a framework so that we can, after we practice the framework, we can release our brain to think about the hard things, which are the domain problem and the statistics, not structuring the meeting. And then there's the content of collaboration and communication. One of the things about communication that we've developed is how to ask great questions, how to explain statistics well to non statisticians, and then finally, the relationship as an aspect of a collaboration is very important. And it comes naturally to some people and not so naturally to other people. So for me, I usually think about the task of the project, and not about the relationship. But because of the asker framework, I realized that actually developing a strong relationship with a domain expert is a goal of the collaboration. And so I focus on building that stronger relationship helps me do better statistics and accomplish the task.
Rosemary Pennington
Well, that's all the time we have for this episode of Stats and Stories. Eric and Olawale, thank you so much for being here today. Thank you. 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.