Skip to main content

Speaking About Statistics: A Conversation with Nancy Reid

By , In , In News

Nancy Reid’s biography is a testimonial to the extraordinary contributions she has made to statistical sciences during her long and fruitful career. Professor Reid is a University Professor of Statistics at the University of Toronto; a Fellow of the Royal Society, the Royal Society of Canada, and the Royal Society of Edinburgh; and a Foreign Associate of the National Academy of Sciences. In 2015 she was appointed Officer of the Order of Canada. She has also received numerous awards, including, most recently, the inaugural David R. Cox Foundations of Statistics Award from the American Statistical Association, presented at the Joint Statistical Meetings (JSM) in Toronto in August 2023. CANSSI members hold Professor Reid in special regard for her impact as the Institute’s second Scientific Director (2015–2019). In an interview with CANSSI’s communications officer just before the JSM event, she spoke about her career, her ongoing relationship with CANSSI, and the statistical sciences community. Below is a condensed version of the conversation.
What drew you to study and work in statistics originally?
I started out in computer science as an undergraduate at the University of Waterloo, but it was in the days when computer science meant programming, pretty much. We did have to take a statistics course, and I found that that spoke to me more than the computer science courses did—they had an interesting blend of mathematics and some computing, and then also whatever particular area you were applying the statistics to, so you could learn a little bit about a lot of different areas.
How would you describe your overall research program for a layperson? What are the main thrusts of what you’ve spent your life doing?
Although I got interested in statistics because I liked all the different ways it could be applied, most of my research ended up being in the more theoretical end of things—so what I tried to do, or ended up doing, I guess, is trying to find the general mathematical principles and tools that provide a basis for building statistical methods to apply to lots of different settings.
How do you think the statistical sciences community has changed over the course of your career?
Well, there’s been a huge change. There’ve been many aspects of this change, but the most striking one to me has been in the last maybe 15 years. It’s suddenly become something that most people have heard about and have some passing interest in, whereas when I started out it was somewhat obscure to the layperson. But I think now there’s a more widespread appreciation that there’s lots and lots of data that’s being collected all the time, and statistics is one aspect of helping to make sense of that.
There’s been a push for diversity, and it seems to me that statistics is maybe ahead of a lot of other fields in terms of diversity. What’s your impression?
It’s ahead of some other STEM fields, I’d say, and there are particular subfields of statistics that have pretty good gender balance, like biostatistics, for example, and statistical genetics; so I’d say in terms of female and male it’s gotten pretty good—it’s improved a lot over the years. In terms of other aspects of diversity, I’m not sure we’ve made so much progress. I don’t think it would be too different from other STEM fields in terms of geographical diversity. The classes now don’t look like my class at Waterloo, but even so, I think there’s still a long way to go.
You’ve received so many awards and you’ve written books and you’ve led organizations. When you look back, what things stand out, what things bring you the most satisfaction?
I think the best part is when you get an occasional note from a student from a few years ago, or from many years ago, that says something like “That course was great” or “I took stats because of you” or something like that. That’s probably the number one, is hearing from students that they remember you and you made a difference. And I have to confess, it’s also quite pleasant to occasionally read a paper from some years ago and think, “Well, that wasn’t so bad; I think I can be proud of that paper.” The awards are lovely, but I think the things you really remember are the people more than the stuff.
You were CANSSI’s second director, and I’m wondering, how did you come to take on that role, and why did you think it was important?
Well, I took it on because they asked me, but probably it made sense to ask me at the time because I was quite involved with the funding landscape for statistical science in Canada and I’d worked on a long-range plan for research in the mathematical sciences. I’d been involved in a precursor to CANSSI that was started 10 or so years earlier, so I think that I was one of the obvious persons to ask, but I took it on partly because the first director, Professor Mary Thompson, was actually my professor at Waterloo, and she made an impact on me the way I hope I’ve made an impact on others. She had done an amazing job, and I felt in some sense we owed her a great debt, but I also thought it was really important for statistics in Canada that this continue to build from where Professor Thompson left off, and I thought there was lots of scope for expanding.
What do you think are the most important contributions CANSSI could make in the coming years?
That’s a good question. I think the most important strength of CANSSI was that it was set up, from the moment it started and even from before it started, to be national, to reach across the whole country, and also to be interdisciplinary. I’ve seen lots of enthusiasm from university administrators and research administrators for interdisciplinary research, but I’ve seen little recognition of how difficult it is to do that type of research. It’s really hard to find funding for it, and it’s hard to be rewarded for it, and I think by building it into the framework of CANSSI from the start, that was a great achievement and a big help to the statistical sciences community. I think that’s a big strength, and we should never lose sight of that.
More generally, where do you see the statistical sciences going in coming years? I’m thinking especially in light of developments like big data and artificial intelligence and machine learning.
We statisticians can be a bit cup-half-empty, and we seem to be perpetually worried that our turf will be taken away, we’ll lose ground to computer science or to other fields of research. But I try to be more cup-half-full about it because I think that statistical science has a pretty unique role in science. It’s maybe a bit more risk averse than computer science, but on the other hand, I think we bring an intellectual effort to the table that’s different from machine learning and different from artificial intelligence, and one way to describe it might be, it may be more focused on understanding a system than on results—which might sound a bit negative, because who doesn’t want results, but understanding where the results come from, I think, is important to statistical scientists, and I think it will always be important in science. So I think there will always be a key role for statistical science.
What advice would you give to a young statistician just starting out?
I guess that that’s a little hard to answer because there’s so many different aspects to it. But I would say, in more general terms, in my 40 years of working, I don’t think I’ve seen a time when it’s been more exciting for statistics. There’s so much going on, there’s so many different problems and so many different aspects of statistics that are of interest—even just in the last four years, the COVID epidemic put statistics front and centre, so I think it’s really never been a better time to be a statistician in. Just find what interests you and run with it. I don’t see how you can go wrong.
Last question. What work activities are you focused on these days?
Well, I seem to be perpetually scrambling to finish things. I’ve spent today preparing the talk for the Joint Statistical Meetings in August. The point that I’ll try to emphasize in my talk is that the abstract and theoretical foundations of statistics continue to be relevant for any number of particular problems that society is confronting, so I guess I continue to maintain an interest in linking the foundational aspects to current issues.