SWB volunteer chat: Bryan Sayer
Sloka Iyengar, PhD, past Director of Marketing and Communications at SWB (and current Secretary) talked to Bryan Sayer, SWB volunteer. This interview was conducted in February 2023.
Sloka: Thank you for your time today, Bryan. Could you talk a little about your professional journey and your association with Statistics Without Borders?
Bryan: I studied statistics in a business school at the University of Texas as an undergraduate and then demography in graduate school at Johns Hopkins University. I am probably more of a demographer than I am a statistician, but I particularly like sampling, specifically, the development of sampling frames. I have been most involved with SWB on sampling projects.
Sloka: What is your non-SWB life like?
Bryan: I retired last year um from one of the government consulting firms in the DC area. We did primarily healthcare consulting for the government with the National Institutes of Health (NIH) and the Agency for Healthcare Research and Quality (AHRQ). AHRQ did a lot of work in digestive diseases for the National Institute of Diabetes and Digestive and Kidney Diseases. We worked on a publication called The Burden of digestive diseases. There is also a survey called MEPS (the Medical Expenditure Panel Survey), which I did a bunch of work on, at different times.
Sloka: Do you miss working on those kinds of projects?
Bryan: I do still occasionally consult, but I don’t miss it, as I have many other things. I have taken up photography again and I build plastic models.
Sloka: How long have you been part of the SWB community and what attracted you to this organization?
Bryan: I did international development work, particularly in AIDS research in East Africa, and knew I wanted to do some more of the international work. I signed up to become a volunteer at SWB a couple of years ago, but I didn’t volunteer for any activities until relatively recently when I had the time to do it, and I wouldn’t have to worry about whether I could get to it or not. I tend to watch out for opportunities that are sampling oriented. Right now, I am working on a project in Afghanistan. The other one was helping with sampling for camps for internally displaced people, and cross-border people. Ground Truth Solutions had requested help as they were doing an in-person survey of migrants and displaced people, and they were asking these individuals what aid they got and what aid they needed in the camps. They were doing it across eight or ten with probably 30 maybe 40 different camps. We were helping them because they (the organization) could not go to every camp, so we helped them group the camps and select ones to visit and send their interviewers to.
Sloka: What were some of the challenges that stand out to you?
Bryan: When dealing with displaced people, we have big thick books with many techniques, but it always boils down to how two things. The first being how much information can you bring to the table to draw your sample from, and two, what resources do you have to do it. Usually, there is not a lot of money to do this work, and not much data either, because these populations are changing very rapidly. In this instance, they had some counts from the Displacement Tracking Matrix from The Institute on Migration (IOM), and that is quite decent, but it will be non-current very quickly. While we did the best we could, it was difficult to get good information. I did enjoy seeing the information from the Displacement Tracking Matrix.
Sloka: Is the survey all done?
Bryan: Yes, as far as I know, the interviewing is completed, and they were going to head to Ukraine next. But I don’t know exactly where it stands.
Sloka: Do you think the work that you and other SWB volunteers did on the displaced individual projects, e.g. its methodology or would it be used for other things as well?
Bryan: One issue in survey statistics (and in statistics in general) is the issue of intercluster correlation coefficient (ICC) which becomes important anytime we study clusters (in this case, camps). One suggestion was that the organization could publish their observed ICCs which could inform other activities, and other people doing surveys of displaced people. We did not have access to the complete data including the interview results; all that we had came from sampling frames, but it would have been useful to publish the entire analysis.
Sloka: In your long professional Journey in the field of statistics and specifically in the field of sampling, are there certain things that you’re excited about? Or perhaps certain things that you that you’re looking out for, or you want to change?
Bryan: My interest is in voting turnout and voters and small area estimation. It is a hot topic right now with Artificial Intelligence and big data. I think there’s going to be a lot of things happening in those areas. While the theoretical stuff is interesting, there are many limitations too when we are dealing with the developing world, which are important to keep in mind.