And The Winner Of The First International Prize Of Statistics Is ... Sir David Cox | Stats + Stories Episode 36 / by Stats Stories

36 ellenberg.jpg

Dr.Susan Ellenberg is Professor of Biostatistics, Department of Biostatistics, Epidemiology and Informatics , Perelman School of Medicineat the University of Pennsylvania, with a secondary appointment in the Department of Medical Ethics and Health Policy. She was the hair of the International Prize in StatisticsFoundation that will be awarded at the World Statistics Congressin July 2017.

+ Full Transcript

(Background music plays)

John Bailer: I'd like to welcome you to today's Stats and Short Stories' episode. Stats and Short Stories is a partnership between Miami University and the American Statistical Association. Today's guest is Professor Susan Ellenberg from the University of Pennsylvania. She was the Chair of the International Prize in Statistics Foundation, that will be awarded at the next World Statistics Congress in Marrakech, Morocco in July, 2017. I'm John Bailer. I'm Chair of the Department of Statistics at Miami University and I'm joined by my colleague, Steve Siff, Professor in the Department of Media, Journalism and Film. We are delighted to be speaking with Susan on our short episode today. Susan, before we discuss the first winner and the reason for selecting this individual, can you tell us a little bit about the international prize, such as, you know, what is it intended to recognize and why was it established and who sponsors it?

Susan Ellenberg: Well, the International Prize in Statistics was established to bring recognition to the leading lights of the profession and also to bring recognition to the profession as a whole. Many other professions have major awards. We know about the Fields Medal in Mathematics and of course there are other important math prizes like the Abel Prize and the Wolf Prize. We know about the Nobel Prizes. We thought there should be a major international prize in statistics and this is sponsored by major statistical organizations - the American Statistical Association, the Royal Statistical Society, the International Statistical Institute, other international societies and we actually hope other societies are going to join in, in the future, in sponsoring this important award.

Bailer: If you could picture a drumroll right now, who was the first recipient?

Ellenberg: The very first recipient, probably to no one's surprise, was Sir David Cox of Nuffield College, Oxford. Professor Cox was a very strong choice and a very popular choice for this first award.

Bailer: Was it a hard process? Was it hard to choose?

Ellenberg : I don't know much about the actual process. The way this was set up is that we have a foundation who oversees how things are done and then there was a selection committee and the selection committee reviewed nominations which came from all over the world. So I don't know what went on with their deliberations. I just know that they came back with the judgment that Professor Cox should be the awardee.

Steve Siff: Is it difficult explaining to outsiders, the importance of the work that leads up to this kind of prize?

Ellenberg : It really should not be hard with the work that Professor Cox has done, because his work has touched just about every area of science. I am a biostatistician. I work in medical research so I'm mostly familiar with the application of the Cox proportional hazards model, which is, you know, I should say Professor Cox has had a very prolific career. He's written 18 books and edited a few others and has published about 300 papers in many different areas. But the Cox proportional hazards model is a seminal method that is used in many, many areas of scientific research and that was really the basis of this award. So when you ask, is it hard to explain, certainly anybody who does science, in almost any area, will probably be familiar with this methodology.

Siff: What is the Cox model, for a layman?

Ellenberg: I'll put it in the area that I'm most familiar with...medical research. If we are looking at treatments and how well they do in preventing bad things, for example, preventing death in the most severe cases, and you're comparing the results of two treatments, what you want to do is to see over time, if one treatment is doing better than the other. And the Cox proportional hazards model allows you to do that very efficiently and accounting for other possible factors. For example, if you are comparing two groups and one group was substantially older than the other, that wouldn't be a fair comparison. You would want to be able to adjust for the age difference and you can do that with a Cox model.

Bailer : When was the first time you heard about the Cox proportional hazards model, as a professional?

Ellenberg: Well I heard about this when it was when it was first published in 1972. I was…I wasn't even a statistician then. I was just starting on my statistics studies and I was working as a computer programmer for Jerry Cornfield who was recently retired as the chief of biostatistics at the National Heart Lung and Blood Institute and was a very eminent world…internationally known biostatistician. I was doing programming for him. He had a research grant and he became aware of this publication like I said it was published in The Journal of the Royal Statistical Society in 1972 and he brought in Jim Ware, who was then a young statistician at the Heart, Lung and Blood Institute and ended up going to Harvard sometime after that and did…had most of his career at Harvard … but a very, very smart young guy … and he brought Jim in to explain this new Cox model to us. We all sat there wrapped up in Cornfield. We all idolized him and he told us this was something really important. And Jim gave a presentation and I remember that my sense was, oh, this is way too complicated. You know, that the physicians that we work with will never accept this. It's too hard. They'll want to stick with the way we were doing things like just comparing the numbers still alive at two years or five years. Well what did I know?! Within a very short period of time any study that you were doing that involved looking over time to see when an event would happen, all the, all the collaborators very quickly adopted and expected to see a Cox Model used in analyzing their data. I think it was really an example of, you know, if you build a better mousetrap, you know, people are going to find it and use it. But that was my first reaction that you know it was just too complicated. People would not go for it but they did!

Bailer: It worked out all right for him then I think!

Ellenberg: It sure did!

Bailer: Well it's been our pleasure to have Susan Ellenberg join us on Stats and Short Stories. Stats and Stories is a partnership between Miami University's Departments of Statistics and Media, Journalism and Film and the American Statistical Association. Stay tuned and keep following us on Twitter or ITunes. If you'd like to share your thoughts on our program send your e-mail to and be sure to listen for future episodes where we discuss the statistics behind the stories and the stories behind the statistics.