'Statisticians need to have a seat at decision-making tables': John Pullinger on his upcoming years as RSS President
- Author: Statistics Views
- Date: 14 Feb 2013
- Copyright: Photograph appears courtesy of Mr John Pullinger
On 1st January this year, John Pullinger took over duties as President of the Royal Statistical Society from Professor Valerie Isham (to see interview with Professor Isham, please click here). Pullinger’s extremely impressive career path has led to his current role of 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 from the merger of the Central Statistical Office, Office of Population Censuses and Surveys and statistical units of the former Employment Department, the policy lead on development of the Government Statistical Service and responsible for international relations.
Statistics Views talks to Mr Pullinger about his upcoming years as President and the International Year of Statistics, and his thoughts on current issues from chairing the getstats campaign, the challenges the Office for National Statistics face, how we deal with Big Data and memories of representing the UK at the United Nations Statistical Commission and European Union.
Video interview in which John Pullinger talks about his upcoming years as RSS President
1. How we deal with Big Data is a hot topic at the moment and at the Big Data Debate held by the Economics and Social Research Council recently, the view maintained was that it has enormous potential to help look at the subtle interactions between data, such as linking clinical data sets for the improvement of medicine and hence better treatment of patients. There is another side of the coin with the existence of these very large data sets and people may draw conclusions that may not be warranted in terms of the quality, not to mention the ethics in dealing with such data. With your background in data service, what do you think can be done in dealing with Big Data and your thoughts on approaching its analysis and challenges?
I share the general view from the Debate that there is a massive opportunity with Big Data. We now have the potential to do things with data we never could before. However, it is still a potential. There are currently no rules out there and rules need to be developed. I am convinced, if we deal with this question, Big Data is a positive step for us.
For me the big issue that has been underplayed so far is data provenance. What we have currently is masses of undifferentiated information spread out over the Web. It arrives just from the click of a search button and apparently answers your question. But the skill we all need to learn is checking what procedures were put in place to ensure this information’s validity and quality. What kind of sampling was put in place to make sure it is representative to the population it was designed for? These basic questions have never typically accompanied such data before. When a principal investigator collects analyses and publishes the data, his/her own reputation is dependent on giving assurance all the way down the chain. The more we have people trying to use each other’s data to get these links and subtleties that are there to be found in Big Data environments, the more it is necessary for the data to be provenanced at every level. This is critical in my opinion.
Another issue is the ethics of using Big Data, particularly in relation to personal data. In the medical field, there is a good tradition in protecting patient confidentiality. But increasingly it is going across social domains from official data sets like the government holds like our benefit or tax records and there is a need for us to be protected.
Let's play with Big Data and see what we can do with it
We hear increasingly more about debates about the kind of personal data that Google and Facebook deal with. The ethical and legal framework backing that data up is still incomplete as we have not had to face these kinds of questions before. Until now it has not been possible to do the kind of matching that Big Data allows. We have the technology and the knowledge. Subject to these two caveats, we should see how things unfold. We need a framework for bad issues to be stopped but without discouraging innovation and creation of new developments. It is hugely exciting and can be greatly beneficial to our lives. The kind of apps we have now are amazing. It has been snowing heavily recently here in the UK and you can now check your journey home via your phone to see if you are going to get held up or not. That is clever use of data that we never could have done before and it makes your life easier. Let’s play with Big Data and see what we can do with it.
John Pullinger on the current challenges the ONS faces and the getstats campaign
2. What has been the most exciting development that you have worked on in statistics during your career?
There are many and I’ve enjoyed every job I’ve done. The most exciting have been when we’ve discovered something new and then that new discovery has improved a decision that needed to be made. One example is during my first ever job as Assistant Statistician for the government’s Department of Industry which I was extremely fortunate to get. It was at the time of what was then called the worst recession we’ve ever had during the late 1970s. It was my job to monitor the investment for the manufacturing industry – all the surveys showed unprecedented reductions and that was not surprising but there was one small cell in our survey that showed strange fluctuations. Our survey was not big enough to work out what was happening but you could tell from the basic data and analysis there was something funny happening in this particular sector. It was not in the manufacturing sector and had not been paid attention to. It was the finance sector. What we were starting to pick up in the survey were the beginnings of finance leasing – companies having a much bigger role in the economy that we then saw growing enormously during the 80s and 90s. Having discovered this, I was taken along to a meeting with the Treasury who were checking their forecasts for the economy in the upcoming period and they were seeing just how far the manufacturing industry was falling off a cliff. They were then able to see that some manufacturers were still able to invest because they were getting their investment through a finance company. That insight would not have been possible without a statistical survey that I as a 21 year old took part in and spotted this pattern, analysed it, produced a report and was able to brief the people who were responsible for informing the Chancellor.
Much later, right at the absolute other end of the scale, it was exciting to be sitting at a desk at the UN building in New York with United Kingdom written in front of me. We were debating the system of national accounts and getting agreement at world level on the statistical methodology for comparing the economic performance of all countries. There is not a day that goes by without someone in the news talking about how GDP went up or down or how are we doing compared with the US or China. It all comes down to that meeting in New York and you got a real buzz from that!
So in conclusion, what I have enjoyed most is being involved in new developments which will help all sorts of people make decisions.
3. 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?
At last year’s RSS conference, one of our chief speakers was Hal Varian, Chief Economist at Google who is famous for saying how statistics will be the sexiest profession to work in during the next 10-15 years. The reason he said that is because of Big Data. At Google, he has had vast experience of bringing together mass quantities of data and turning it into new opportunities. One of the things he talked about was Google Translate which you can easily find on your tool bar and translate to any language you like. That functionality is all based on statistical algorithms that either did not exist or are not as comprehensive as they are now. So that is the area for development – it’s the applications, tools and methods that enable you to get information out of Big Data. What the statistical community needs to do is to cross-fertilise ideas that are emerging from, for example, the corporate world of Google, the medical world of genomes and DNA sequencing, the physical world of data coming from the Hadron collider. We are creating vast amounts of data and statistical thinking and methodology is the way of making sense and decisions out of this data. The excitement is bringing it all together and coming up with new methodologies that enable it to be brought to use in all walks of life.
So big data is my first area of innovation. My second is measuring progress. Even at the UN meeting in New York that I mentioned, there was the feeling that traditional methods for measuring performance were not as adequate as they need to be. Now it is becoming very urgent since the financial crash. Levels of employment are puzzling considering economic output. We need numbers that help us make sense of what is going on. I have no qualms about the quality of the figures being produced but there is a serious need to stand back and think what is it that we are trying to measure that will help us in decision-making? Two years ago, the PM announced a decision to measure happiness and well-being in statistics – how do you measure that? What is the purpose of the government in the wellbeing of the people? We then take these statistics and analyse them and they can in turn guide policies. During the 20th century, the GDP helped inform so many improvements, reducing serious poverty in many countries including the UK but the world is a different place now. Statisticians need to have a seat at these decision-making tables as once you understand the political imperatives, you can think what the statistical questions are and what are the methods to answer them. Our challenge is to bring these worlds together to guide progress over the next 100 years.
What the RSS can do is promote the value of the professional statistician.
4. What do you see as the greatest challenges facing the profession of statistics in the coming years?
We need to distinguish professional statisticians from someone who uses statistics to support an argument and purely purvey the numbers for their own use. We have a chartered statistician accreditation led by the RSS but it is not a well-known qualification and it needs to be. One of the frequent complaints I receive from MPs is that they have the latest information from a pressure group showing that a particular policy must be pursued because the evidence is there but the evidence has not been done with proper statistical principles and it does not have that professional stamp on it. What the RSS can do is promote the value of the professional statistician.
5. Are there people or events that have been influential in your career?
Statistics is very much a people business and we all have people we look up to. The person who has influenced me most is Claus Moser. As an icon about the importance of statisticians having a relationship with decision-makers, I think one of the most creative relationships over the past 50 years was the relationship between Claus and Harold Wilson. As the government’s chief statistician and the UK’s Prime Minister at the time, they worked very well together and this enabled statistics to be mobilised not just in the Prime Minister’s office but in all walks of life. They symbolised what can come out of such a partnership and Claus’s ability to make that connection was wonderful. More recently, I’ve had the joy of getting to know him as a very wise person and he is always able to stand outside of an issue, think what the real question is and care about the importance of statistics making an impact on people’s lives.
The late Sir Roger Jowell was a pioneer of survey research, collecting information about social attitudes and later creating the European Social Survey enhancing our ability to understand what people’s attitudes are to put alongside the harder measures. He stands out for me as I got to know him and then his work. It was a great tragedy when he died last year. He is very fondly remembered by me and many others in the profession.