John Pullinger is the current president of International Association for Official Statistics (IAOS) and finished his five-year term as United Kingdom National Statistician, Head of the Government Statistical Service (GSS) and Chief Executive of the UK Statistics Authority in June 2019. There, his role was to safeguard the production and publication of high quality official statistics by all departments, agencies and institutions in the UK. On appointment he described his role as to, “mobilize the power of data to help Britain make better decisions.” He also has represented the UK internationally in EU, UN, OECD and other forums. He was both Chair and Vice-Chair of the United Nations Statistical Commission (UNSC).
+ Full Transcript
Rosemary Pennington: Did you know that younger adults report feeling lonelier than older adults? Are you curious about why life expectancy is flatlining? Or why the number of trees on your street affects the price of your home in the UK? If so, there’s a wealth of data to help you understand these and other issues. Much of that data is collected by the National Statistical Agencies in countries around the world. The work of such agencies is the focus of this episode of Stats and Stories, where we explore the statistics behind the stories and the stories behind the statistics. I’m Rosemary Pennington. Stats and Stories is a production of Miami University’s Departments of Statistics and Media, Journalism and Film, as well as the American Statistical Association. Joining me in the studio are regular panelists John Bailer, Chair of Miami’s Statistics Department and Richard Campbell, former Chair of Media, Journalism and Film. Our guest today is John Pullinger. Pullinger is the current president of the International Association for Official Statistics and finished a five-year term as UK National Statistician, head of the Government Statistical Service and chief executive of the UK Statistics Authority in June of 2019. That is a mouthful John, thank you so much for being here today.
John Pullinger: It’s a pleasure.
Pennington: So, I’m just going to start off very broadly asking you why we should care about official statistics?
Pullinger: Well, I think without official statistics, democracy can’t function. How do we decide who to vote for, unless we can check out the claims that politicians are making about what’s going to happen to the economy, or society, or the environment? How can we judge whether they’ve been successful? How can we look at whether their manifestos actually stack up? Official statistics are the lifeblood of democracy.
John Bailer: Wow that’s a bold statement. Wow.
Pennington: Conversation over.
Bailer: I mean you started at a really low level here John; that’s exceptional. That’s great and you know it’s- you know how far you’ve come from chief statistician at ONS to being on Stats and Stories. I mean this is -
[Laughter]
Bailer: No, we are delighted to have you join us John. You know, one conversation – we both were just at the UN Stat Commission meeting last week and someone made this statement that I thought was really intriguing- it’s that the idea that statistics is a human right. That seemed like a really strong and bold statement; do you have some thoughts about that or why someone might make that assertion?
Pullinger: Yeah, I mean if you think what are- what is needed for a human being to function properly, I think there is a fundamental right to be informed about what’s going on in your society. Unless you really know what’s going on around you in a way that you can be confident is valid, you’re kind of wandering around in the dark. I mean you’re as disabled without statistics as you are without the other human rights. And when the UN was creating the fundamental principles of human rights, they were thinking about what are these things? What is it we need to be effective as human beings in society? And that was way back in the day. I mean, now, when data is really the ubiquitous commodity in the world, the right to be informed with trustworthy data stands out even more if we’re to be good functioning human beings, not just in the UK and the USA, but in other countries around the world.
Richard Campbell: So I’m intrigued by that very inspiring first statement you made and it’s similar to something that I saw at an interview you did- I think when you stepped down from the UK position recently, Statistics are a pillar of our society supporting the decisions we make at home and at work as individuals and collectively. And the first thing I thought there is we could substitute journalism for statistics. So, and I think that’s part of what the spirit of our podcast is, because I think for Rosemary and I, we’re interested in the job that journalists do reporting data and reporting statistics because I think the work that you do gets translated by journalists. So, my question is how good of a job are journalists doing? And are there other ways that we could team up? Since I think that both statistics and journalism, I agree with you, are sort of central to democracy working well.
Pullinger: Well, the message I used to give to statisticians in my organization was your job is not done when you’ve published the statistics, your job is only done when it’s reported accurately and fairly, and understood by the audiences you’re trying to reach. And if that’s part of your job, a key part of you is a journalist, whether that be a journalist on TV or whether it be somebody who is working in social media with a particular community that to you. So, journalists are really friends of statisticians in this place, but they need help. I mean a lot of them need a lot of them- support to be able to understand data and we need to spend time supporting them. I mean the initiative that I would particularly commend is the Royal Statistical Society’s award for statistical journalism. Every year you would get the best journalists in the room with the statisticians that are reporting this and celebrate what’s good online and on more traditional mediums. Journalists are the way in which statistics gets faithfully translated to the public, or not.
Bailer: You know that’s a- the American Statistical Association has a similar type of award and we just featured one of the winners of that award on a recent podcast, and you know it seems like that would be aspirational for pretty much any country’s National Statistical Societies to think about having that type of award and recognition for outstanding reporting of statistical information in their communities. I really liked in one of the articles that you had written, you were saying that in an opposed truth world statistic could provide a central public service. And one of the statements you made within that editorial piece was perhaps our real challenge is to take our statistics off the page and find ways to listen and connect with those people who’ve been left perplexed and disappointed by experts. And I was wondering if you could give an example of a case that you thought was very- where that was done very well, very successfully, and perhaps another example where not so much, where it was not successful?
Pullinger: Well, I guess one very poignant example for me was when I was working on poverty statistics in the UK, and one of my colleagues had come up with this conclusion that a particular housing project in London was the worst place to live in London. And they produced the numbers and they were going to go and take them out to a meeting of people in this place, and the statistician was very anxious about going out there because he was worried about the reception. And he got in there and did his presentation and there was complete silence and he was really worried they were going to turn ugly. But the community leader who was in the room with him said no, what they have realized that they now have a voice, they know this is the place with the worst crime, the poorest education outcomes, poor health. But now the statistics are actually showing it they can take those into the local authority and make their case. Now that to me is the kind of powerful story of statistics connecting with individuals and enabling them to assert their rights. Links together these various . Now where I think it’s been less successful, and here’s another UK example and it’s an example that I’ve been widely quoted of during the referendum for the UK to leave the European Union, politicians were going around talking about the state of the economy, and at one public meeting someone was presenting the GDP and someone in the audience got up and said hey, that’s your GDP now, that’s not my GDP. And what we were saying in a profound way was that these macroeconomic numbers were not speaking to that person in their community. And I think that’s where statistics in the UK particularly needs to respond, needs to make sure the statistics, it is producing to informed debate do accurately reflect the reality on the ground. In the UK school systems, the macro does not connect very well. You have to have numbers to connect with individual places particularly, particularly to those places that are left behind, and particular communities whether you've been working in particular sectors of particular ages of people, or disabled people whatever it is unless the numbers are actually relevant to them, why should they recognize them?
Campbell: You know, one of the things that you’ve said is that one of the ways, for instance, to counter fake news or misrepresentations of data science is- and this is a quote from you- the strong use of narrative as well as numbers, which I think resonated, those of us at the table who are journalists. Do you when you speak to groups you talk about- how do you tell a story about complex data? I mean, I know how journalists- journalists are good at telling stories, not so good with numbers and statistics. So, there's- I'm interested in that strong statement where you you mentioned narrative and the importance of narrative.
Pullinger: Well, I mean, unless we communicate our statistics, we’re wasting our time because the point, the value we get from statistics is . And as human beings we can understand narrative better than we understand numbers. So, we need to translate the numbers on the page into a voice that people will recognize; nothing we can't learn from. We employed within the office of National Statistics a number of- particularly digital journalists who could really create pictures, create stories, create narratives that eventually hit the spot with people. Using video- that is the medium by which people interpret the world, so statistics needs to be using that kind of media. But it doesn't come very easily especially for statisticians, because we are trained to be much more numerically oriented, obviously. But we spent quite a lot of time in the releases in the UK that we have produced, so crime. When we get data coming in from the police, we get data from victim’s survey, things like violent crime, we get data coming in from hospitals. And the team and they’re coming up and it's just work and work. What is this really saying? What’s the collective? You can only really tell that with the story, turn it into something that’s a narrative like crime or domestic violence. And actually, that can help statisticians say something in a few words that will be really hard to get to the level with numbers alone. And often the best thing you can say about numbers is don’t change them. Don't forget to show the difference is not statistically significant; that's always going to be misinterpreted. As soon as you start using those words- if you say but it's gone up a lot more than it did last year, people will get you. This amount is not statistically significant and they’re a bit lost. What do these numbers really say? You can explain it to your friend, the one you're going out to dinner with. If you’ve actually understood the numbers. If you can’t, you haven’t really understood your numbers.
Pennington: As you’ve been talking, I’ve been thinking about a TED Talk I saw David Putnam do several years ago, where he was talking about actually the relationship between press and politics in the UK, and he raised this point, and I teach it a lot in my journalism classes when I talk- we talk about ethics, but this issue of inflaming versus informing, and how as journalists one of our missions has to be deciding if we're going inflame or inform people, and that has to be- we should be informing, right? But you have to sort of carve a road and it seems like, you know, you mentioned this this GDP statistic that was being used in the lead up to Brexit, how do- how do statisticians navigate that space of having this information, like about violent crime? And figuring out how to communicate it in a way that is informative, rather than sort of being packaged in a way that will inflame people.
Pullinger: Well, first of all you mentioned David Putnam and I’ve worked very closely with him, but my previous role was working in Parliament and how you communicate democracy and David shared a mission to do that. So, I learned a lot of my craft from him so that link is a very nice, well made one. But David- obviously is a film director, so you kind of taking that kind of methodology from film. You think of like film works, you work very hard at it, you have multiple takes and lots of ways you’re trying to frame it. You boil down months of footage into maybe a few minutes of film, statisticians need to realize this is something he needs to learn. When I first had to go on TV, I said to myself couple of days having a really hard to time with some really brutal people, and I gradually learned to; and that would be my message to my colleagues. I'll also say it's not for everybody as well, and not every statistician is going to be a brilliant communicator. And that's fine, too. You need to know you need to get some friend who is a communicator to sit by your side and help you get your message out.
Pennington: You’re listening to Stats and Stories and today we’re talking with John Pullinger, President of the International Association for Official Statistics.
Bailer: So, John you talked a little bit about how you became a better communicator and one of the things that you had done to challenge yourself to improve in that regard. I think that’s an interesting question. As I -as we as we think about training the next generation, whether it's of journalists are statisticians. You know the journalists that work with the statistics or the statisticians to better communicate. Do you have kind of other words of wisdom or suggestions for helping us to prepare that next generation?
Pullinger: Well, I think you need to think of the roles that people are in and then design a curriculum or competency framework, depending on what situation you’re in, that really enables them to be successful. And so, within the government system service in the UK, we had a competency framework, which was very heavily dominated by technical elements and that's kind of fair enough, but it wasn't enough. And so, my conclusion was for every person who is going to be successful as a statistician in the government service, there needs to be a communication? So, it’s actually to define them and work with communications experts in government to think about the competence of communication systems. Now, the interesting thing is having designed that with the statisticians, you them become very heavily taken up by policy officials, other people working in other disciplines within government, who realize that they need to have the statistical confidence to be a good communicator, and there was a kind of sweet spot that was- that level of understanding and appreciation that sits between substantive and statistics. And I think that is a message for people working in academia or business or third sector of whatever else it is, but there needs to be level of subject matter or communications for the statistician to be effective. But there also needs to be the statistical competence for the person working in the other discipline or the other function to be effective too. And we need to, as statisticians, take responsibility for both.
Campbell: You've said when misrepresentation occurs in fake news, let's just call it what we see- whether it's from politicians or whether journalists get a story wrong, you say that it needs to be challenged publicly. Can you give an example of that? How you would actually do that? I'm imagining generally statisticians don't do that, right? How is that done? And can you give an example of how that might be done?
Pullinger: Yeah, now firstly the fundamental principles of official statistics, specifically assert the right to pull out misuses of statistics. In the UK, what’s happened since we had the legislation that's been in force since 2008- we have had a succession of examples where poor use of numbers has been called out by the statistics , up to and including the Prime Minister, various different prime ministers on several occasions. And the kinds of issues where there's been a public statement that this challenged the use of numbers that the level of UK the level of the national debt, which kind of mattered at the time we were going through a period of fiscal consolidation, Flood defenses was another one with the government not spending sufficient money on flood defenses where their numbers did actually stack up. Education performance would be another one; knife crime is perhaps the most salient one. Where ministers were making claims about knife crime that just weren't substantiated by the numbers. So not only was that called out, but as a consequence the calculation of that ministry into the Office for National Statistics in order to restore trust in the numbers. So, we have our kind of long tradition now going back 10 years or more of there being an expectation that there will be a kind of official body whose job is to test whether these claims are valid or not.
Campbell: So, in the in the UK how good are the journalists at reporting it when you call out this, you know, the misrepresentations? Are you really dependent on them to get that information out there?
Pullinger: Sure. I mean it is a great story for them, isn't it?
Campbell: Yes, you would think it would be, yes.
Pullinger: And so they do tend to, I mean they were they will like, often someone- they will make public that they have referred an issue the statistics authority, and then the journalists will be bringing up every day, well have you decided ? And that's a tricky one because some of these facts are quite difficult to validate, and really make sure you get to the bottom of the claim can be very hard. And on some topics the numbers are really difficult, and would be an example. And you need to make sure you are making a judgment about statistics, rather than a judgment about political discourse. Numbers can be very tricky. We don't feel we have enough time to adjudicate it and get some real clarity about what’s right and wrong before coming to a conclusion. Sometimes I think the conclusion is well you can't really say very much about this. It can be a bit disappointing, but again you've got a really think you’re doing something quite important in the democratic space here.
Campbell: So, let me let me ask the other John siting at the table here, do we have anything comparable in the US? That’s that sort of official, where there is a requirement to call out misrepresentation?
Pennington: And why aren't you doing it?
Campbell: Yeah. Yeah, very good.
Bailer: I think there’s fact checking lots of different places, but I'm not sure if it at the official level, John are you aware of it?
Pullinger: The fact checking movement, I think is a very powerful one. And Full Fact in UK. There are similar bodies in the US, across Africa, in various other countries, and I think a civil society movement that has got a real good track record on that, in some ways is even more powerful than an official movement. Creates a following and a resonance with audiences. And certainly, in the UK I had an extremely and fruitful relationship with Full Fact and we kind of simple number to make how best we could validate the numbers, and certainly they develop a very strong link into our website, so, they'll be able to fact check in real time. So that means when politicians are making claims live on TV, there would be a feed that would let you fact check as the story was going on. Sometimes in the context of an interactive TV show, the claims will be tested during real-time. That's when you start to really be able to stop people making false claims because they know before it had a chance to hit the spot.
Bailer: So John, you were just at the UN Statistical Commission meetings last week and that that was part of the origin of things like the fundamental principles of official statistics has been a product of the UN and colleagues of yours and you. Can you talk a little bit about what the fundamental principles of official statistics are? And why is that an important statement to have been done at this transnational level?
Pullinger: So, the fundamental principle celebrated their 25th anniversary in 2019, and they were initially in 1994 and were triggered by the fall of the Berlin Wall. What they were trying to do was to set some principles for the countries of the east. They developed statistical systems to serve their populations in the new democratic environment. What was great about that was you had this country who was still very fresh in the idea of democracy thinking about how we design an information system for our citizens that will help create and foster better government. So, the key things for me is the fundamental principle number one, which is that statistics are an essential element of a functioning democracy. And they must make the best practical use of utilities in the principle. So, the key point is to be useful. So, it's creating numbers that are relevant to the issues that your particular country is facing at the moment. And that's the way the UN has adopted it, so as the UN has moved forward, it's considering those issues where the current evidence needs to be approved, and statisticians facing individual countries and the world ought to be addressed properly. So, in the economy for example there is a continuing process stating what an economy actually means to ensure that emerging sectors are really well understood and measured and put into the right context where in other sectors it might be less important.
Pennington: John, given your position and your interest in official statistics, is there a story that maybe, across-the-board, that reporters miss when it comes to the numbers that are produced by agencies? You know, there's a lot of reporting on GDP, on economics numbers, on poverty, on unemployment, but is there something that governments are collecting data on that reporters should be covering that they're not?
Pullinger: So, the big agenda in the UN over the last four or five years has been sustainable and when this was being considered by the UN there was a lot of attention on the economy, social change and on the environment. And traditionally, we kind of thought well that kind of covers everything, doesn’t it, looking at those three things. But there was a very great consultation done across the world, statisticians generated data and what do you actually care about? And the thing which came from that was that one of the things people care about is trustworthy institutions. They don't like corruption. They don't like institutions that are unfair; institutions that leave some people out, or work for elite rather than the population as a whole. So, it’s something that was newly put into that sustainable development goals this time. There has not been any measurement framework before. It’s what's now called gold sixteen, and that covers issues like peace and security and a whole range of issues, and we don't measure those very well because they've not been in this kind of group. But that came out from listening to people. Saying, what do you need? If you want companies to succeed, you want to make sure you and your family are not really left behind, you want to get rid of corruption, you want to improve transparency. So now we have a series of indicators in the framework that capture those things, but very few countries have yet been able to find every last method of measuring them. But we will, and we’re working on some of these; in some areas there are some good practices. I’ve mentioned crime statistics, the fear of violence is also in part of living in a peaceful society. There are countries that measure victimization quite well. The UK and Mexico are two good examples. So, we are now trying to export what we know to others, other countries are now picking up other aspects of criminal behavior. We’re working with organizations like Transparency International to understand how you get measures of transparency of government. But once you’ve got an indicator framework at that level, then you can start what matters on the things you can always measure. So that’s the best example I can think of.
Pennington: Well, John that’s great and thank you so much for being here. That’s all the time we have for this episode of Stats and Stories. Stats and Stories is a production of the Miami University’s Departments of Statistics and Media, Journalism and Film, and the American Statistical Association. You can follow us on Twitter or Apple podcasts or other places you can find podcasts. To share your thoughts on the program send your email to statsandstories@miamioh.edu or check us out on statsandstories.net and be sure to listen for future editions of Stats and Stories, where we discuss the statistics behind the stories and the stories behind the statistics.