"Let's get back to science and make our governance work": Prof. John Hinde on becoming President-Elect of the International Biometric Society
- Author: Statistics Views
- Date: 16 Jan 2013
- Copyright: Photograph appears courtesy of Professor John Hinde and NUI Galway
During last summer, John Hinde, Professor of Statistics at NUI Galway was elected as the next President for the International Biometric Society. This year he steps in as Vice-President before becoming President in 2014. Statistics Views met up with Professor Hinde last month to ask about his plans for the Society, the International Year of Statistics and his own career. In celebration of the International Year of Statistics, Wiley have a virtual issue in biometrics freely available to read here.
You continue to teach at NUI Galway as Professor of Statistics. As a university professor, what do you think the future of teaching statistics will be? What do you think will be the upcoming challenges in engaging students?
Statistics is not always the most popular subject in universities. The on-going challenge is to motivate students that statistics is useful and can be transferred to science and medicine, etc. We need to think about the courses we teach and make them more interesting, accessible and challenging. It is increasingly hard to do this due to the lack of resources – it varies from country to country but certainly in Ireland at the moment, we have staff shortages and people are under more pressure to teach larger groups of students. This isn’t always the best way to teach statistics. One wants to teach in a more focussed and applied way to smaller groups.
At the other end, there is the challenge to teach more specialist statisticians who will eventually fill posts in research and departments. Students may not always see the career opportunities and the public perception needs to change – statistics has a lot to say and can contribute towards solutions of many important problems.
I think that many people who teach undergraduates are well aware of these issues and we will carry on doing the best we can in the current climate.
What do you think it is about R that has been so appealing to statisticians today as the preferred software to use? Why has it been so widely adopted over the past few years?
One simple answer – it’s free! It is an extensible system and it’s easier for people to make available new developments they have done. When you work on a new method, you can make the R package available to the community. But that was also true of other systems before R. Today’s interview is being held at the Royal Statistical Society, a supporter of the development of GLIM, an early extensible, albeit limited, system. More general packages, such as GENSTAT in the UK and SAS in the US, also allowed user extensions, but these were more expensive and perhaps not so easy to use.
As R was free, it was picked up by everybody and it is relatively easy to use, especially for more standard methods. It’s a nice environment in which to develop new methodology and people also have super tools now so that they can easily embed the R code together with text, so you can have journal articles where the analysis presented is supported by R code. People can see what is going on, follow these methods and extend them to their own work. Also, the range of guides on statistics with R has rapidly increased over the years. There were systems that were close but R came along at the right time.
Do you think that statistics undergraduates and postgraduates starting out today are under more pressure to publish and to obtain grants than when you were a student yourself?
I think that is definitely true. There has always been some pressure to publish but it is now more extreme. Young statisticians need to get papers published before they move on to a post doc position where again they need to keep up a good flow of publications and start applying for grants themselves. The pressure does not stop when they get a university position. The expectation will be that in five years’ time they will have a significant research profile. There is not much pause for reflection. There is perhaps too much focus on getting published rather than a concentration on the research itself and a missing of opportunities where the research could be improved and extended to be of benefit to others.
Do you have any advice for students considering a university degree in statistics?
It’s an exciting and interesting subject where they get to play in all sorts of different areas. Almost whatever they are interested in can be examined by the application of statistical methods. It’s diverse in its outlook. Students also gain experience in substantive areas such as research in biosciences, so that they can then appreciate how applied research interacts with statistical analysis rather than from just the mathematical end of the spectrum where one can treat the subject in a more abstract manner. Statistics is often taught as part of mathematical sciences and it’s not clear that this is always best and perhaps we should be encouraging a more diverse range of topics for students so that they can then get into exciting subject areas that they had not contemplated before.
Over the years, how has your teaching, consulting, and research motivated and influenced each other? Do you continue to get research ideas from statistics and incorporate your ideas into your teaching? Where do you get inspiration for your research projects and books?
One always tries to reflect research into one’s teaching, particularly on a more advanced level. That may be simply be in terms of applications and consulting that one has been involved in to illustrate how statistics can be applied to all sorts of interesting problems. It is of course much easier to integrate current research at the post-graduate level.
What has been the most exciting development that you have worked on in biometrics during your career?
I don’t think I’ve made a major contribution. I have chiselled away at the edges! I’ve contributed with certain types of models with applications in biometrics and I have worked on random effects and mixture models. I think what I have enjoyed most has been collaborative projects where we have put together advanced methodology and come up with interesting answers. Like lots of statisticians would say, I think, I’ve enjoyed working in the collaborative projects and the time I worked at Lancaster University was a very privileged period – for 5 years we worked as researchers, did some consultancy and had a programme of visitors and that was a very exciting time for creative work and innovation. I have tried to create a similar atmosphere at Galway and we’ve built it to a reasonable size with postgrads and post-docs, and have visiting professors and advanced courses as well.
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 biometrics during the next few years?
The most important recent developments are that there are now tools to help you work on complex model frameworks which offer both power and flexibility and have changed the whole modelling environment. Years ago, you thought such models would be interesting but there was no way you could apply them – now it is perfectly doable. For me, the future is probably more of the same.
What do you see as the greatest challenges facing the profession of statistics in the coming years?
One of the main challenges is the volume of data now produced and how do we apply statistical methods to find what scientists are looking for in these huge datasets? Bioinformatics and biostatistics with the impact of genomics are hot topics and challenging areas where we will see interesting new work coming in.
Recruitment is probably the main one! There are plenty of jobs for statisticians but not enough statisticians so we need to work out where we are failing to communicate to young people in seeing statistics as an exciting career opportunity.
Are there people or events that have been influential in your career?
Working with Murray Aitkin in Lancaster set me on the path I now follow and I have continued to work with Murray over the years.
I have also worked with several people in Brazil, which came about by a chance encounter meeting with Clarice Demétrio, the current President of the IBS, with whom I’ve now collaborated for many years. She is at the University of São Paulo and they have lots of applied biometric problems that gave me the impetus to go back to work in biometrics.
They both changed the path of my work over the past 20-30 years.