Last summer, John Pullinger stepped down as Royal Statistical Society President to succeed Jil Matheson as the UK’s National Statistican.
Pullinger’s extremely impressive career path led to his previous role of House of Commons Librarian and Director General of Information Services for the UK Parliament. Prior to this, he was Director of Policy and Planning at the Central Statistical Office, in which he was the project manager for the creation of the Office for National Statistics.
Statistics Views originally talked to Mr Pullinger about his upcoming years as President when he first started in the role back in January 2013. Pullinger now reflects on our past interview, what he did achieve during his Presidency and how the past year has gone in his new role as National Statistician, including the seven, yes, SEVEN, stages of interview it took to be appointed!
1. Due to your new role as National Statistician, you stood down as President of the Royal Statistical Society and we last spoke when you were just starting out. During your Presidency, membership increased, the online presence of the society expanded and your term also saw the International Year of Statistics, which included very well-attended talks in partnership with Ipsos Mori and Kings College London. What are your memories when you look back on your time in this role?
It was wonderful to be part of a society at a time when it was rediscovering itself. What I found most satisfying about the long-term strategy reviews that I had inherited as President was that we were going back to the original roots of the charter. There was a campaigning mission thinking about statistics in society in the kind of broadest way and going back to the role of some of the writings of the original pioneers like Florence Nightingale on what statistics can be if used effectively. So it was a lot about rediscovery and working with people on the cutting edge and using new data sources and techniques to understand new problems.
I carried out fairly straightforward statistical work and there was a greater confidence amongst statisticians to take that work forward to government and schools. I hope that statistics in schools has had something of a renaissance. Being President at a time when statisticians were becoming more confident was very satisfying.
A real highlight was seeing the growth of the Young Statisticians section. At the last two RSS conferences, the YSS (in very bright T-shirts) have been everywhere. But it’s not just the brightness of the T-shirts; it is the kind of brightness that they are giving the profession and a profession which is growing. This profession is attracting great minds which is again very satisfying to someone at the other end of their career who can see that the profession is in safe hands for the future.
2. During our last interview, you said that the RSS can promote the value of the professional statistician and that a statistician should have a seat at every table where decisions need to be made. ‘Big Data’ is ever more on people’s lips now than it was when we last spoke in January 2013. Do you think that the reputation of statistics has improved within the last three years?
I think there is now a very clear recognition that if you are using data to make decisions, you want to be confident that the analysis of that data has been done in a professional way. In the same way that happened in the 19th century for medicine when the profession of doctor was better recognised – someone who had been through rigorous training and certification – this is also now happening for the professions of accounting and engineering. I think that we are getting to the stage where it becomes just as unthinkable for someone to do a critical data analysis without a professional statistician behind it, as it would be for a bridge to be built without a chartered engineer or chartered accountant. Professionalism matters in a world where decisions, particularly critical ones where lives could be saved, depend on the work being done to a very high professional standard.
3. We know that governments and learned societies need to do more to raise awareness of statistics in the general population and stop people being afraid of numbers. What do you think the RSS and Office of the National Statistician can do to help raise public awareness of statistics and prevent the misuse and abuse of data?
In terms of the RSS, this year marks the halfway point of the getstats campaign which was launched in 2010 as a ten-year campaign to turn around the perception of statistics in society as being difficult or irreverent. I think there has been tangible progress such as greater recognition in politics and in the media that it matters to get statistics right and that people who play fast and loose with statistics need to be found out. Organisations such as Full Fact, that didn’t exist five years ago, are now very much working at the forefront, challenging the misuse and abuse of statistics.
In terms of this office, it does have a particular role to play. We need to ensure that the message of our statistics gets across in quite a crowded market. Immigration is a good example whereby as ‘Perils of Perception’ found, the public thought that there were substantially more immigrants in this country than there actually are and that is because the media is full of stories about immigration. We have collected the authoritative official statistics and we need to present those in way that cuts through the jargon. Any of that reporting is prefaced by very simple straightforward information about what is really going on so that everyone can understand.
This year we have specifically launched a website for the public (http://visual.ons.gov.uk/) which is about taking statistics down to local area level so the public can look at topics and data of relevance close to them in a language that they can understand.
I think by using the traditional application of the statistician, particularly in understanding bias, together with extraordinary technical possibilities that are coming through the field to help us, will help us to better understand data.
4. With an educational background in geography and statistics from the University of Exeter, when and how did you first become aware of statistics as a discipline?
The time I really understood just how much of a difference statistics could make was during my first year as an undergraduate where I took a course on social trends. Data for the course was from this very office and that is when I realised that good statistics pulled together and presented in ways that people could understand could actually influence the way in which people interpret the world. I thought right there and then that this would be an interesting career to follow.
5. Are there any exciting developments that you are looking forward to working on?
I think the challenges of unconventional data sources would stand out for me. I think that there is a lot of hype about Big Data. At the moment, many people believe that you can replace traditional methods simply by getting more data out there and because there is more of it, it must be better. I think by using the traditional application of the statistician, particularly in understanding bias, together with extraordinary technical possibilities that are coming through the field to help us, will help us to better understand data. Some of the things that we are working on in this office include supermarket data, for example, to look at price changes and trends in buying. Such data is sourced entirely from scanners and other areas in the supermarket may not have been considered. With Twitter, you can form these cloud-like images of most commonly used words but the questions are not being asked such as, Twitter users are concentrated in one particular area of the country, so what is happening to those who are missing? Is there somehow a lack of voice for those who are not online or who are not using a source that is data-rich?
Statistical techniques that help us understand the quality of data are going to be very significant and even more precisely, something that is going on in the public realm this year is looking at the measures of success for countries across the world from 2015-2030 in terms of goals for sustainable development. Some of those goals are in areas which we are used to measuring – classic demographic measures, for example – but some of them that matter to us do not have a good statistical matrix so we need to look at information around good governance, peace and security. There are areas where the data you can get hold of that is very uncertain, such as displaced people. We need to understand the patterns of these people and there are certainly new data sources that will help us understand but tackling such data can be very hard. Trying to think of new methods to harness new data that will help us to tackle future world problems is going to be pretty rich pickings for the statistician in years to come.
Copyright: Image appears courtesy of UK Statistics Authority