Tim Harford is an economist, journalist and broadcaster. He is author of "Messy", and the million-selling "The Undercover Economist". His newest book “The Data Detective” was released in the U.S. and Canada earlier this month. Harford is a senior columnist at the Financial Times, and the presenter of Radio 4's "More or Less", the iTunes-topping series "Fifty Things That Made the Modern Economy", and the new podcast "Cautionary Tales". Tim has spoken at TED, PopTech and the Sydney Opera House. He is an associate member of Nuffield College, Oxford and an honorary fellow of the Royal Statistical Society. Tim was made an OBE for services to improving economic understanding in the New Year honors of 2019.
Episode Description
Our lives are framed, every day by data and statistics, though we may not always be aware of that fact. Helping us make sense of this universe of data is the goal of many an economist, statistician, and journalist. It’s also the focus of this episode of Stats and Stories with guest Tim Harford.
+Full Transcript
Rosemary Pennington: Our lives are framed, every day, by data and statistics, though we may not always be aware of that fact. Helping us make sense of this universe of data is the goal of many economists, statisticians, and journalists. It’s also 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 as usual are panelists John Bailer, chair of Miami’s Statistics Department and Richard Campbell of Media, Journalism, and Film. Our guest today is Tim Harford author of the books “Messy,” “The Undercover Economist,” and “The Data Detective: Ten Easy Rules to Make Sense of Statistics” released this month in the United States and Canada. Tim is a senior columnist at the Financial Times, and the presenter of Radio 4’s “More or Less,” the series “Fifty Things That Made the Modern Economy,” and the podcast “Cautionary Tales.” He’s an associate member of Nuffield College, Oxford and an honorary fellow of the Royal Statistical Society. Tim was made an OBE for services toward improving economic understanding in the New Year honors of 2019.
Tim Harford: Thank you, it's it's great to be back on the program,
Pennington: So one of the things you do is that you write this regular column for the Financial Times where you often are sort of helping people make sense of various kinds of data or stats. Could you talk a bit about sort of how you decide what data you want to write about and sort of how you figure out what the story is that you want to tell your readers.
Harford: Sure. I think it's really changed over the last year. So if you had asked me this question a year ago, I would have said, I noticed things in the world around me that I think, bring up interesting questions in social science I'm trained as an economist but I'm interested in science, psychology, statistics, interesting ideas, maybe I read a preprint that's come out of the National Bureau of Economic Research or a study that's come out of a think tank. And I pick it up and that's that would be the basis and I would start with that work and start by trying to connect it to other things I know about and translating it into a language that that my readers understand. But more recently, of course, we've all been trying to understand the coronavirus pandemic. And so, from the start, I was trying to get on top of the data, understand what that data was telling us, and explain to my readers some of the key issues and answer to the most important questions so early on, for example, the central question was, well, how, how dangerous is this really, and it could it be that there are lots and lots of undiscovered cases. If so, it's a lot less dangerous than we think. And it's a lot more widespread than we think and we'll be at herd immunity by last spring. Well, turns out we weren't, but it wasn't ridiculous to think that in March, so just exploring what to do and don't we know from the data and the central thing is, I have questions, and I want my questions answered. And if I think it's interesting and I want to understand this then I believe that my readers will also find it interesting and I suppose that is the common theme before and since the pandemic started.
Bailer: Yeah, I think you've just identified one of the things that motivates our selection of episodes for the podcast. I mean it's if there's someone we want to talk to and we think it's an interesting topic it's a great place to start.
Harford: I'm very flattered by that.
Bailer: Well I very much enjoy your writing, I, I have to confess that I, I love how quickly you're able to turn this around. I mean, Yeah and and being able to do this on a weekly basis I mean this is nightmarish in terms of, oh Lord, I've got to get another one out but. One of the recent articles that you read one of the columns that we said, were, what have we learned from the great working at home experiment, that was the title of the post and the article the column that you wrote, and that really was one that I thought wow I mean that that drew me in so, so can you recap sort of what some of the insights that that you kind of explored and and you came to and looking at this question.
Harford: Yeah, well I'll try and remember, although say John you didn't he didn't give me advance notice you asked. One of the things about being a journalist is since I wrote that piece I think was probably the first piece published this year we're having this conversation. Just after Valentine's Day. And so, that piece was then published six weeks ago so that's six columns ago it was written I think before Christmas. Okay, so it's written eight columns ago I've also since then mentioned. I've written. I've presented six episodes and more or less, and I've presented eight episodes of how to vaccinate the world. So just to give you a sense of, you know, when I think this is pertinent to your question because we're talking about this tournament like, you know, I'm on to the next thing. It is actually a very interesting topic, it was motivated, partly by my own reflections on my own experience, and it's fascinating. Economists out there Nicholas Bloom, who's done various bits of work with various people trying to put data on our intuitions so date him, one of the things that he did early on, but way before the pandemic was to run a randomized controlled trial with a Chinese travel company. And they were experimenting with working from home because they thought it would save them on rent and save rent. And there was some surprises that emerged from that study. And it was an interesting opportunity for me to talk about, I don't think I use these words I'm pretty sure I didn't use these words but both internal and external validity of that study than, and because I'm, you know, I'm not necessarily delivering a statistics class I didn't use the words but the fact that because it's a randomized controlled trial, because there's a very high degree of confidence that what we started there is real. But it also raises questions. It was pre pandemic. it's in China. It's a call center. Therefore, what does that tell us for example about people who work in retail, in America, or in accountancy in Europe. Maybe not a lot, or maybe something and I was able to, to gesture towards those questions without using all the technical terms.
Campbell: So, Tim I'm interested in the relationship. This after all is called stats and stories the relationship of telling stories about data and how you see that relationship, the importance of sort of narrative, and explaining data, and just more about your process, how you sort of think through the structure of this, of the data you're trying to explain.
Harford: It varies very much depending on the different medium that I'm using. So my classic Financial Times common probably doesn't have a strong story to it, it's 850 words, I need to have some kind of thesis, some question I'm answering, something I'm arguing, or at least something I'm exploring. And there may be a little then there might be a little protagonist there might be a little anecdote, a little hook, but mostly I'm going through what does the evidence shows us. So to give you the one that's top of my head which hasn't been published yet. I wrote one recently about what is the cost of all these schools being closed for this time. And I began with a little story about my son who's nine. Being in tears on a zoom call because he couldn't answer the questions because the teacher had scrolled the questions up too fast and he couldn't get her attention. So you get the little personal interest and people go oh yeah I can imagine being a nine year old boy not being able to see the exams. The questions, but then I'm into okay what do we know, and we've got the Center for longitudinal studies here in the UK, doing quite serious work on who is actually doing the homeschooling is that the mums are the dads. Everyone says everyone thinks it's the mums. What does the data show? That's actually what they chose. Yes, it's the bumps but it's good to actually be sure rather than where we got these assumptions. So for mothers of primary school aged children, that's 11 or younger. It's five hours a day on average. So, there's a lot of people doing full time clearly. for fathers it's two hours a day which is not nothing, not trivial, but it's clearly a lot less, and for secondary school that's 11 to 18. Mothers two hours a day, father's one hour a day, but we're able to, I didn't linger on the details but I'm able to pull that out of, out of report link to the report, people can follow it up, and so throughout that column I'm saying, okay here's data, we've got on mental health. Here's data we've got on achievement of children how much they learning, I can get the links I can explain it and I'm really just giving people a hopefully a fresh summary of the literature, but an alternative is to say, for example, with the cautionary tales podcast, here's a, here's a story I want to tell you, it absolutely starts with the story. The story is an excuse for me to talk a little bit about social science, social science might be three four minutes, and the story might be half an hour. So that's a totally different format. And one of the things I love about my job is I get to do both.
Bailer: I really like how you've integrated into your, your columns in terms of how you present material can have some surprises of how you frame, frame summaries, an example that I really struck me when I was reading you wrote the article, COVID-19 How close is the light at the end of the tunnel. And as I was reading along and in this article you said COVID-19 is overwhelmingly a disease that spares the under 60 s. And I really liked the framing of that, instead of saying, talking about the disease, where you have this incredible susceptibility and fatality rates much higher fatality rates in an older the older community, that gets it, use us flipped it so I love the turn of phrase I love the focus and I'd like to hear what some of your thought process that led you to that was. --
Harford: Thank you. I'm glad you noticed that one so that was a it was a response to, so I wasn't conscious in choosing that turn of phrase, but I was very conscious of what the turns of phrase I was rejecting yeah so initial. What I initially said is it's a disease that overwhelmingly kills the elderly, which is true in that the elderly are overwhelmingly more at risk than the young, but it's not true, in the sense that I mean even if you're 90 years old. You get COVID you're more likely to survive than to die. So to say it overwhelmingly kills the elderly is is ambiguous and potentially highly misleading. So okay, well, what do I say that, well you know and I sort of played around with different things but in the end I said well, hang on the and, and also I was framing it as a piece of good news. So the good news is the vaccines are highly effective. Even if you only have a small number of doses delivered well why is that well it's because the elderly are most at risk and there aren't very many of them so you don't need many vaccines to protect them. But, but then again well if I'm trying to present this as good news. And the good news is this disease kills these people, well that's no good that's not good news. So instead of twisting around it and in the end I found the well I forget the exact turn of phrase but this idea that it spares the young look if it's past the young. All you need to do is protect the elderly and and the vaccine is already doing a tremendous amount of good so that I think by a process of elimination is where I ended up.
Campbell: Very good. I want to speak up for the elderly here I get my second Pfizer vaccine Monday, so. Oh, no. In Ohio, 87% of the deaths are over 65. And so we're and you know I think our governor here is doing a pretty good job. I worry that the vaccine won't be there because I have to drive 40 miles to get it that's where I found the closest place that would give me an appointment so it's an interesting for sure.
Harford: Yeah, well the evidence does suggest that the the first dose is probably pretty efficacious for, but, but, I mean that is something that I've explored in my radio show and in my columns and that question of how, what is the wisdom of postponing the second dose for these vaccines that turns out to be a really interesting question, and we economists I think I may be more used to more us to talk about this sort of thing than the hardcore epidemiologists because they're all the evidence based medicine, people are saying, Yeah, we've got no data on what happens if you, if you don't give the Pfizer vaccine after three weeks. And the economists, the economists again yeah sure we don't have any data but we, we can make an educated guess, like, well, we shouldn't guess. Well, yeah, but if you spaced out the second dose you could vaccinate twice as many people in the short term and, and 1000s of people are dying. Every week I mean See I don't know what the latest data in the states is but it's been 10,000 a week and there was a time when it was more many more than 10,000 a week. You could save a lot of people's lives. You know, you really not willing to speculate. So that was a very interesting question to discuss the first versus second dose question or postponing the first dose question and to say what do we know from high quality evidence. What do we what is an educated guess. And what is completely unknown. And how do we balance those things and I wrote a column, about that. Then I came down sort of saying, In the end we need to run more trials, but while we're waiting for more trials to be run. I think it's reasonable to postpone the second dose, but hopefully my readers will have been able to read that column and say, Oh, interesting given what you say I disagree with your conclusion. That's what I would hope because I don't see it as my role to persuade people very often I see it as my role to inform people and they can reach their own conclusions.
Pennington: You're listening to stats and stories and today our guest is to Hartford whose latest book the data detective was just released earlier this month. Tim before we started this conversation you mentioned since we're talking about vaccinations. This new project that you're working on how to vaccinate the world. Could you explain what that is and sort of what the goal of that is.
Harford: It's a really interesting departure for me so I realized that the really the coolest radio shows and podcasts out there involve a panel of four people discussing the issues. And you can think of any.
Bailer: Wait wait wait let me count here. Yeah.
Harford: So I just looked at myself. Yeah, we should do that. So, clearly I wasn't the only person to think this so the BBC decided we want a show about vaccines, and the format they picked was me and three people who actually know what they're talking about. And the three people who know what they're talking about vary from week to week epidemiologists economists who studied vaccine incentives vaccine manufacturers bioethicists science journalists and so on, and each week we pick up a important issue so that this week, we will be working on this question of variants, and what do we do, what's the interaction of these variants with vaccines, and we'll, we'll talk about we know what is a variant How is it different from a strain or a mutation. I don't know but by the time we finish recording the program I hope I will know how quickly can we adjust the vaccines, what's the process for modifying these vaccines, what are the obstacles or the potential snags, how worried we should be so we have half an hour discussing those sorts of issues. We put some listeners' questions to the experts, the conversation goes on probably for an hour and then my long suffering producer will edit it all down. And that's really interesting for me because my other radio work, I had a podcast called 50 things that made the modern economy which is very scripted it's an essay by me, illustrated by music cautionary tales, again, it's a it's a tightly scripted story with actors contributing more or less is a program about stats, and it's, it's, again tightly scripted pre done interviews that are tightly edited you might talk somebody for half an hour, and you come out with two minutes of material.
Bailer: Oh wow.
Harford: So this is, this is different, it's, it's a slightly compressed live interview edited down a little bit to make it more digestible. So, very interesting for me as a journalist. A lot of it's in the preparation. Now what questions, am I going to ask you guys forgotten more about this than I will ever know. But it's it's a fun format, because you're able to respond in an authoritative way, and also hopefully in quite a friendly way to the issues of the day. The challenge of course is, you never know quite how the guests are gonna work out, and some are amazing. And when we want them every week and some guests like, you know, we, they were great on the phone and then when we actually got them on the show they they stiffened up a bit and you do your best.
Pennington: It's reminds me of infinite monkey cage, which is one of my favorite podcasts where Brian Cox and Robin Ince interview various scientists but they usually include an a comedian. As part of the panel right shows like I feel like I've learned so much that I that I should have understood as an adult from listening to that because it does feel like very accessible and and friendly and I feel like I'm getting an education without sort of being lectured,
Harford: It's an absolutely terrific show and, and Robin actually Robin very kindly wrote to me today's review of how to save the world just the other day so he's very much in my good books he's, he's great at the difference I suppose with infinite monkey cage and how to vaccinate The world is partly that they, they have this more comic element. They've got a comedian as one of their hosts and the comedian is one of their guests, but also that they will pick a more general topic like let's talk about what we can learn about human history from the fossil record and they can record that two months ago, three months ago six months ago and it's still fresh. Whereas, we are trying to follow the new so that's our that's the extra discipline we have, which in some sense, in some ways, makes it difficult more and more difficult but in some ways makes it easier since you don't need to work so hard to tell people why they should be interested in, like, Oh, this is exactly the question I was hoping someone would answer this.
Campbell: If you're interested in kind of a broader question. How do we reach or get to in the states we have about a third of our adult population that believes the election was rigged. You know that believe that don't believe in getting vaccinated, that wear ignorance is kind of a badge of honor. Follow our former presidents lead, just trust your gut. So, how I think that's the sort of larger problem going forward is, is the sort of anti science anti data waves that are out there I don't know if it's as bad in the UK I know it's pretty much everywhere now and we have a lot of authoritarian leaders who trust their gut. Do you, do you have any strategies about what do we do about it, how can we do about this sort of a larger problem about disinformation.
Harford: It's a huge issue and it is one of the reasons why I wrote my book the data detective. I suppose the first thing to say is, we still try and get our own heads straight first. So people ask me a lot about how I persuade other people, and my response is always well let's just make sure you're thinking straight first. Once you're happy you're thinking state but then maybe you can get on with trying to get help other people think straight but we're all capable of fooling ourselves. You said you suggested that there was an anti science or anti data movement and I think that that is that can be over. It can be overemphasized. What I would say more is that people are interpreting the science and the data very much through their own lens it's and those lenses are highly influenced by their own social groups and by their political affiliation and so on. So, for example, a republicans tend to be very skeptical about climate change and the threat of climate change. And I think most people, adopting that position especially more educated, who are more skeptical about climate change, by the way, the more educated they are the more wider this polarization is which I think is interesting. They would not say I'm anti science, they would say well I you know, these are the scientists that I respect. This is the data that I should pay attention to and I and they would find some reason to say that well your scientists are wrong and your data is wrong. So it's a more complex thing than simply rejecting science, it's, it's more being very selective about the voices that you listen to and the data you pay attention to. And I'm my book has various pieces of advice. The most important I suppose is simply to, to begin by being curious and looking at the data as a way to inform you about the world, rather than looking at the data as a way to win an argument. And we all know we all want to win arguments. We all want to persuade other people that we're right to win them over to our point of view. But that's rarely an attitude that helps it may help other people sometimes but it's never going to help you be smart. You are always going to be smarter if instead you're putting yourself in the position of trying to understand how the world works.
Bailer: I find it very much a challenge to think that you know what what will change anybody's mind what would change my mind about something, and you know, that's that's not a trivial task and, and, you know, the one thing this show is has taught me or reinforced for me is that you know it's not data doesn't convince will not convince people to change.
Harford: Yeah. One thing I try to do is that I think does is intriguing, is to ask people instead of having an argument with somebody, and trying to persuade them to ask them to explain their position, then not even to justify their position but just explain their position. And what data, are you looking at, what is it, what is it that, what is it that makes you think that vaccines are not safe. What What is, what is your source of information there just explain that to me. And sometimes when you open up like that to people, they will end up talking themselves around to a more sensible point of view, because it's simply faced with this person who's listening and paying attention, they realize, Oh, actually. Now I come to explain this maybe I didn't really understand it as well as I thought. But even if that's not the case at least you've shown them some respect. And that is a more constructive way of arguing, and trying to dismiss what they say and, you know, and convince them. I say this, it's really hard. The I, I find it very very difficult to be patient with for example. Well, I have no patience with anti vaxxers. I do find it. I struggle with people who are vaccine hesitant, for example, I think the vaccine these vaccines are amazing. The world has been burning for a year and now we have the solution and you're like, you're not so sure. But this doesn't help doesn't help to lose your cool, and so I try to be calm and try to be respectful,
Pennington: there's a point in your book where you sort of remind readers of the importance of sort of gut check like, you know, how is the data making you feel. Yeah, and sort of how are you emotionally responding to the data as you're trying to make sense of it why I mean that sort of feels kind of semi really like but why is like understanding our emotional response important in this situation,
Harford: yeah i mean that's the very first chapter of the book and I think it's the one that's most surprising, in the context of all your classic guides to statistics and loads of great guys to statistics I don't think any of the others have this piece of advice. But when you reflect on the experience of the last few years, you reflect on the last two presidential elections, you think about people's views on climate change or vaccines, or in my home country Brexit, you just think about it and you realize we're emotional creatures. Human beings are emotional, emotional, and we believe what we believe, largely because of who we trust what our friends believe and what our what our instincts what our emotional and what our values are what our emotions tell us what we want to be true, what we expect to be true. These are huge influences on the data we choose to pay attention to and the data we choose to reject. And that's true even for scientists as his famous talk by Richard Fineman pointing out how long it took physicists, to realize that Robert Milligan's estimate of the charge on an electron was not quite right. And it's because every time they came up with something that agreed with Milliken they were like, Oh, yeah, of course, and every time they came up with something that disagreed with Milliken. We're not so sure about that. And that scientists and no one has passionate views about the charge of intellectual capital punishment or gun rights. Right. So, the reason I say look notice your emotional reaction is because we all have them, they influence our thinking. You can't suppress it. You shouldn't ignore it, you should notice it. And once you've noticed it counted to three taking a deep breath, maybe you can then go back to the original tweets that you were about to retweet or block headline you're about to discard or share with everybody. And you're now in a position where you're thinking a little bit more clearly.
Pennington: Well that's all the time we have for this episode of stats and stories Tim thank you so much for joining us again.
Bailer: It's my pleasure. Thank you. Thanks Tim
Pennington: Stats and Stories is a partnership between Miami University’s Departments of Statistics, and Media, Journalism and Film, and the American Statistical Association. You can follow us on Twitter, Apple podcasts, or other places you can find podcasts. If you’d like to share your thoughts on the program send your email to statsandstories@miamioh.edu or check us out at 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.