David Morganstein is currently the Vice-President and Director of Westat’s statistical unit, where he has worked for more than 35 years.
Morganstein directs a unit of 65 statisticians. He specializes in the design and application of surveys and systems of evaluation, quality control, statistical analysis and estimation. He leads research and development tasks that improve or
enhance Westat’s survey and census projects and has developed best practices for several key survey processes.
Morganstein, a fellow of the ASA and member since 1972, was President of the ASA during 2015 and previously served the association as treasurer and Vice-President, an at-large member of its board of directors and a member of several ASA committees. He is an elected member of the International Statistical Institute and an instructor in the Joint Program in Survey Methodology. He earned a bachelor’s in electrical engineering from Purdue and a master’s in statistics from the University of Michigan.
Alison Oliver talks to David Morganstein about his Presidency of the ASA and his career.
1. When and how did you first become aware of statistics as a discipline and what was it that inspired you to pursue an MA in statistics at Michigan after studying electrical engineering at Purdue?
Somewhere in the third year of my engineering studies and with some summer work, I realised that I needed a career that had more to do with solving common problems that people had and that would benefit them in society in some way. I didn’t see as clear a path in engineering as I did after meeting a mentor by the name of Andy Anderson. Andy was a sociologist and he was interested in mathematical models that tried to understand human behaviour. I took many sociology classes from Andy and that led me to apply to Michigan for a program with Professor Leslie Kish on survey sampling. He had tried to create a program at Michigan called ‘Mathematical Sociology’ and in subsequent years, the program unfortunately evaporated. A prerequisite for getting into the program was obtaining an MA in statistics which I subsequently did.
2. You served as President of the ASA during 2015. What were the highlights?
Travelling to universities and talking to students has been a big highlight. It recharges the batteries. You can be in a career for a whole lifetime and it’s good to find a way to get reenergised with excitement about what we do and the excitement as seen through the eyes of someone early in their career.
Preparing for JSM 2015 was another highlight and it was an opportunity to speak to the members and I think another huge plus is the enthusiasm and the spirit of the volunteers in our Association. There are a lot more people who volunteer than we realized. Steve Porzio and I considered this and we estimate there are about 2000 volunteer positions when you consider all the committees, chapters, sections, meeting and publications.
3. What were your original objectives when you took on the role?
Trying to establish mentoring on a much broader base that is routine and adopted by some chapters or sections to serve their members’. I think that there is a huge hunger amongst the younger people especially to want to feel connected with an individual and with the Association, so certainly mentoring is an entrée to doing that.
One initiative that I wanted to achieve and I did was establishing an annual Outstanding Mentor award with the Board. I think that kind of recognition is important. Another initiative is to have docents at the JSM to welcome first time attendees and help answer any questions. At JSM 2015 we had maybe 1500 of the 6900 who attended for the first time. We want them to come back. We checked attendance at previous JSMs and maybe 20% returned within a four year period. We would like to have more so having a docent welcome them and being able to answer questions has helped achieve that. We can also help with journalists with whom the ASA has made major steps. We assist them to report material that reflects proper statistical science. The more we can collaborate and the more journalists think of ASA as a resource is what we are trying to build.
4. You are also a Fellow of the ASA and have been a member since 1972. How do you think the Association has evolved over the years and adapted to the changing needs of the statistical community?
One of the biggest set of changes is to broaden ourselves by adding a conference on statistical practice and touching into a need of many of our members who are very focussed on practice. I think we serve the members on theory and practice well but there are many who are primarily practitioners first and that conference addresses their needs. We address the training they need. We are increasingly such a global and international society and during my presidential address at JSM 2015 I mentioned that for a decade the majority of grad students in US stats departments come from outside the US. So how do we help them and help to meet their unique set of challenges? We now provide that training at JSM so those I think are important shifts.
5. You are currently Vice-President and Director of Westat’s statistical unit. Please could you tell us more about your every-day role?
We have a group of about 70 survey statisticians, survey methodologists and analysts. It is a kind of a jigsaw puzzle in the sense that Westat works on many different projects in many different areas. We are not like other organisations which are much more stove piped where you work within your own channel. We are highly integrated and our statisticians work on projects in every subject area. An important part of this is trying to keep people both active and busy and not overloaded, so I spend a great deal of time checking in with people to assess their workloads. Other managers and I works together on how we can help people to grow through formal education outside of Westat on job training. We have a very good seminar programme so my job also involves finding good speakers and connecting us with the Washington Statistical Society. A lot of my role involves ensuring everything is working well and that people feel that they are challenged by the work they have. It’s the Goldilocks question, do you have too much work or too little work?
6. You specialize in the design and application of surveys and systems of evaluation, quality control, statistical analysis and estimation. What are you focussing on currently and what do you hope to achieve through your research?
One of my most challenging projects is a longitudinal tobacco survey where we are fitting the second wave of households that are screened and involved after enrolling a year ago. There are a lot of technical challenges. A great deal of what Westat does is one off studies. We are often creating studies on the fly and putting material together for the first time. This tobacco survey has some technical aspects to it and is an interesting technical problem.
The big question for me is, how do we take burgeoning amounts of data which aren’t designed by a survey, and rather than throw away our sound traditions, how do we instead meld them together? There are a few people working on this and discovering valuable aspects.
7. What do you think the most important recent developments in the field have been? What do you think will be the most exciting and productive areas of research in statistics during the next few years?
Everyone will say Big Data as a first topic. Federal agencies are trying to do more with less and asking suppliers to be innovative when they actually mean ‘do the same thing but for less money’, which is a challenge! The big question for me is, how do we take burgeoning amounts of data which aren’t designed by a survey, and rather than throw away our sound traditions, how do we instead meld them together? There are a few people working on this and discovering valuable aspects. We have been good at putting things together. For example, say you have two surveys or two probability samples but how do you take something that has far clearer properties in terms of what it represents with something that you just don’t really know much about and yet still try to get something valuable out of it – that is a real challenge for our association. Then the whole issue as the amount of data increases, some aspect of what we do routinely such as inference changes somewhat because things that you call standard errors shrink to nothing. The problem is that it is not about standard errors anymore, it is more about the bias and representativeness. We have known forever that is much harder to estimate and measure.
8. What do you see as the greatest challenges facing the profession of statistics in the coming years?
Staying relevant. Again back at JSM 2015, I talked about not crossing our arms across our chest and saying we are Rodney Dangerfield who doesn’t get any respect. We engage and reach out, we make partnerships. I think the Washington Statistical Society is a great model. They have teamed up with a data science group in the same area. Each website reflects the other and they schedule activities together. We have found partners in maths and science in general in the past, but now we need to do the same with the IT world and those parts of Big Data that we don’t think of as our speciality but if we team together, the result will be stronger.
9. What has been the best book on statistics that you have ever read?
Too many to pick out but the writings of my mentor Graham Kalton have been very valuable and also those of another mentor, Leslie Kish, whom I used to refer to as the James Joyce of survey sampling text, as he would write in a kind of stream of consciousness style. My colleague Sharon Lord has an outstanding textbook on survey sampling.
10. Here at JSM 2015 you are giving the Presidential Address. Could you please tell us about your theme for the lecture and what points you hope to bring across?
We need to continue to address how to make us a more diverse community and how to serve the changing nature of our membership which includes the increasing numbers of people who come here to the US. Getting minorities and young women into our profession is also important. We need the best people from every walk of life, from around the globe.
11. Are there people that have been influential in your career?
I have several mentors and this goes all the way back to high school. Mrs Ruth Bauer was my maths teacher for three years and she was the kind of teacher that for our graduation gave us a copy of Siddhartha and said, “Study things beyond math and live a life of meaning.” That was early on inspiration.
Another two mentors are Ed Bryant who founded Westat and Joe Waksberg who joined Westat after retiring from a long and distinguished career at the US Census Bureau, whose idea of retiring was 20 more years of working at Westat! The two of them were born in 1915. Ed was born in a log cabin on the Wyoming plains without water and electricity, whilst Joe was born on the opposite side of the world in a town in Poland. As luck would have it, they ended up spending the last 25 years of their lives together. Each year around summertime they have this same conversation. One had a birthday in June and the other in September. The one born in September would say to the other, “I’m sure glad you are still alive and that there is someone here older than me.”