Liberty Vittert is currently a Visiting Assistant Professor at Washington University in St. Louis and on leave from her position at the University of Glasgow as the Mitchell Lecturer. Her current statistical research involves using facial shape analysis to help children with facial deformities. Liberty is a regular TV and Radio contributor to many news organizations including BBC, ITV, Channel 4, PBS, and FNC, as well as having her own TV series on STV (ITV).
+ Full Transcript
Rosemary Pennington: We are a wash and data and information. Almost every day, new stories are published telling us we have a one and one thousand chance of catching some disease, or that eating something we love might increase our chances of getting cancer by some percentage. It can become overwhelming trying to understand what those numbers mean. Communicating statistical information in a clear and contextualized manner 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 Statistics Department, and Richard Campbell of Media, Journalism, and Film. Our guest today is Liberty Vittert. Vittert is a Statistician and a visiting Assistant Professor at Washington University, St. Louis, as well as an ambassador for the Royal Statistical Society. She’s also a media personality, producing pieces for outlets like Fox News, NPR, Popular Science, and The Houston Chronicle. To top it all off, she hosts a cooking show on Scottish TV and was named one of the coolest people in Scotland. Liberty, thank you so much for being here.
Liberty Vittert: Thank you for having me. I’ll also just say that a dog was on the list of coolest people in Scotland. So, you know, it’s not – I mean, I’ll take it, but it’s not [LAUGHTER]. Maybe I’ll beat the dog next year. [LAUGHTER]
Richard Campbell: Liberty, we’ve been doing this show for four years now, and I think you’re probably the coolest statistician we’ve had on the show. John’s pretty cool, but he doesn’t have his own cooking show.
John Bailer: No, no.
Vittert: I’ll take it. I mean, it’s not a high bar with that status [LAUGHTER]
Bailer: Oh, Liberty! You’re killing me! You know, come on, we got ASA sponsorship here.
Vittert: I take it back. I take it back.
Bailer: You can’t – don’t go there. Listen, we’re friends with RSS. I mean, look, you’re giving us really a bad rep.
You’ve had a fairly eclectic journey. You studied at Cordon Bleu, in addition to MIT, and the University of Glasgow. How does someone who researches facial shape analysis, become a TV presenter in the UK?
Vittert: I think by being really annoying and dumb luck, all at the same time. You know, if you bug people long enough, they just want to get rid of you, so they say yes. You know when you’re finishing your PhD, it’s pretty miserable, and you’re procrastinating as much as humanly possible –
Pennington: It’s true.
Vittert: And I saw one of these people cooking on TV and I thought, well, you know, maybe if I annoy them enough, they’ll let me do it.
Bailer: So, you’re saying you started with cooking, not with the stat bits on the news?
Vittert: Exactly, exactly, and that was my original thing was the cooking show, and then I sort of – there’s a lot of girls that want to have a cooking show, so I figured maybe I’d take the statistics and see if I could anything with that.
Pennington: So, you were actually including statistics in the cooking part of the program?
Vittert: Yeah, that would be a very generous way of saying it. A little bit, a little bit. Statistics, maybe adding and subtracting grams might be the equivalent of that.
Bailer: I guess you could have been doing some estimation. That’s about a tablespoon.
Vittert: Exactly, I like – I think this is the kind of statistics we need to teach students, the tablespoon estimating.
Bailer: So how did that transition? Let me ask you this, what was the first statistics bit you did on the news?
Vittert: Oh, that’s a really good question. Oh, I know what it is. I went on Scottish television to talk about the lottery. You know every year there’s some big jackpot somewhere, and all of the sudden it becomes news. So, I think the first time I went on was to go talk about the lottery, and I mean, to show you how few people watch my cooking show, the producer for that - even though it was the same channel - had no idea that I had a cooking show on their channel. So, you know, maybe there’s not exactly a good crossover between news and cooking shows.
Campbell: So, I have a line that I use when I teach journalism students that comes from one of the co-authors of a book called, “Elements of Journalism”, and the co-author is, Bill Kovach, who was head of the Nieman Foundation at Harvard for a while. He says that the journalist’s job is to make the significant, interesting. So, do you feel that’s also responsibility of statisticians?
Vittert: That’s a really good question. I think that there’s a line that statisticians have to walk between being significant and interesting. Every once in a while, you see something really sensationalized, and I understand the idea behind it, because it gets people to read it, but it’s a really fine line that we need to walk. However, if we’re not able to explain the statistics in a way that is impactful upon the individual – maybe that’s a better way of saying it than interesting – but impactful to the individual, then why on earth are they going to listen to anything we have to say and even look at the significant results.
Bailer: So, when you think about the kind of, both, the research that you’ve done and then the reporting that you’ve been involved with, has this changed the way that you write about research that you do?
Vittert: I think that my goal with writing about any research I do is, you have to be correct. There’s no – that’s first and foremost, but at the same time, everyone’s still a person whether you’re reading a research paper or you’re walking your dog. Everyone’s still a person and wants to understand how something can really impact them. So, if we just write down the numbers without the story behind it, without the way that it really affects someone, I think it’s almost meaningless, and I think it really lessens our ability, as statisticians, to communicate with the public.
Pennington: I’ve been really impressed in watching some of the clips from Scottish TV, Liberty, with the way you are able to communicate complicated information, in these really, kind of, bite-sized packages. There’s a lot of training that scientists and researchers go through to try to teach them to be better experts for TV and media, right, because we need quotes and we need soundbites. How do you, when you’re prepping to go on a Scottish news program, how do you, sort of, figure out how you’re going to present information, and what’s, sort of, going through your mind when you’re trying to answer a question and make sure its impactful in a soundbite?
Vittert: I always think about my dad. I know that sounds like the strangest thing in the world, but my father thinks education is useless, that academics are dumb, fake news is everywhere, and I always think about – you know, you get that thing that they always teach you in one of those courses like, are you talking to a child to your aunt or to your grandmother? I think you just think of a person who is not interested in this stuff, and how do you make it interesting to them. So, I always figure that if my dad wouldn’t like it or wouldn’t find it interesting, that I’m just not going to say it. And I’ll practice on him all the time. Every talk I give, every show I do, I always practice on him, and I think that’s – you can’t practice on a colleague, because they get it. You need to practice on somebody that doesn’t understand it and isn’t, frankly, that interested in it. So, my dream is that one day he’ll tell me he has no suggestions, and I will not be holding my breath at that moment.
[LAUGHTER]
Campbell: I’m imagining you do this with students too. I’ve heard in, I think, in one of your TED Talks where you talked about how scary numbers are, how afraid people are. We certainly have a lot of journalism students that are afraid of numbers. So, can you give us an example, maybe, of how you use that, Dad technique, on students to kind of get them revved up and interested in data and numbers?
Vittert: Absolutely! The students are the best. They’re the hardest audience. So, any time I have a new example or something interesting, I will always bring it to the students. So, every hour of lecture, I’ll do one or two real-life examples of misleading statistics or something that’s really, sort of, shocking, and we’ll always pair it with something that we’re doing. So, for example, when I’m trying to explain probability, or relative risk, I’ll use this one story from – I think it was last summer – that French fries are going to kill you, and it was, you double your risk of death if you eat French fries. But really, what it was going from is, the risk of dying, if you’re a 60-year-old man - no offence to either one of you guys - [LAUGHTER] the chance of dying if you’re a 60-year-old man’s one in 100. And if you eat a zillion French fries, all of the sudden it doubles to two and a hundred. So, you have a hundred guys that get to eat French fries all day, every day, for their whole lives, and instead of one dying of them, two will, which, I mean, if I got to eat French fries all that much, it might be a risk I was willing to take.
[LAUGHTER]
Bailer: Yeah, OK, I’m not thinking about French fries now. So, I’m curious, what’s been the hardest story you’ve ever had to report? What’s been the one that your dad fought back the most about?
Vittert: You know, I’ve started doing these, sort of, opinion editorials, which are a little bit more complicated, because you try to keep your opinion out when you’re doing statistics. And so, when I try to do that, sometimes – especially if something’s a little bit political – people end up using your numbers and trying to say whether you’re political one way or the other. So, I recently did an article on the cost of the wall. So, I tried to estimate the cost of the wall that is potentially or not going to be built. And it was trying to completely keep my emotions out of this issue, and really just try to fairly estimate how much is this thing really going to cost. I think that’s the hardest, is when you really start to try to really remove your emotions from something that is highly emotional, and it has enormous implications.
Bailer: You know, another thing I was curious about is, is thinking about being someone who grew up in the US, who ended up transitioning to reporting on statistics in the UK. What’s been some of the hardest, kind of, stories to tell over there, given your origin?
Vittert: I think that it’s a different culture. People think that because we speak the same language that it’s all the same, but it really is a different attitude, it’s a different belief system, it’s just – it’s very different. On the other hand, people always tell you to be so careful, because a lot of people are unhappy with America right now, and all the political issues going on, but really, people are people, and so telling the story there, or telling the story here, my whole goal is always to find that little nugget of information that will allow me to make the statistics that I’m trying to explain, feel something to people, make them understand it in a way that really means something to them, rather than just some random story about statistics. And so, I don’t find it to be that different when you boil it down to the people or people.
Pennington: You’re listening to Stats and Stories, and today we’re talking calculations, communication, and cooking with Liberty Vittert. Liberty, I watched your TED Talk, and I thought it was really interesting in the issue of, sort of, talking about the wrong population when we are put on certain statistic. And you, when you were talking about that, you talk about the coverage of OJ Simpson’s trial, and then you talked about the coverage of the – oh my gosh – eating the bacon sandwiches, and how, particularly with the bacon sandwiches, right, the way that story was covered was really misleading. And when misleading the audience to really understanding what that study was saying, what advice would have for journalists, who are covering complicated statistical stories, to ensure that they are actually communicating what the study says and not necessarily what they think it says?
Vittert: Ask a statistician.
[LAUGHTER]
Vittert: That’s the easiest answer. The one thing that statisticians, I really think we need to do as a group is, not become the data police, and always calling out people and saying you’re terrible. But on the other hand, if we don’t do it, who is? So, especially with the Royal Statistical Society and the Ambassadors, they are there to help. They love getting phone calls about this stuff. That’s what we’re there to do. And I believe the ASA has a version of that as well. And really, it’s asked a statistician. If I was going to do something on journalism that I didn’t understand, I’d ask a journalist, which I do all the time. So, I think, really, the key is to ask someone that really understands this stuff.
Campbell: Are there examples that you can cite of things that you see repeated by journalist’s mistakes that they make, that would help us in training our students, our journalism students?
Vittert: I think the biggest error I see, and I see it every day – and believe me, I love Daily Mail, but the Daily Mail health section is worse about it than anyone on the planet [LAUGHTER] – is this idea of relative risk versus absolute risk. You know this, one in 100 – you’re doubling your risk of death, but really that means you only go from one and a hundred to two and a hundred. It’s that idea that if you have the risk of getting a disease is one in a billion, you can triple your risk and you’re still -- chance of getting that disease -- is still only three and a billion. It’s one of the simplest topics to explain, and I think it’s the one that I see over, and over, and over again.
Campbell: You know, your training was in math and statistics, are there things that somebody training in statistics should take in college that you would have found useful once you got into the work world? At some point you learned to write stories, and that’s not even easy for a lot of our journalism students to do. So, is there something students should be doing now, statistic students? I usually ask the reverse question.
Bailer: You know, I’m on the queue for that one now.
Campbell: So, what do you think?
Vittert: Well, I think – first of all, thank you about the writing stories. I’m not sure I do it that well, at the moment, so I always room to learn. But I really think that we teach students statistics – statistic students, statistics. We teach them how high-level statistics. And I think a lot of times, the basic material, this quantitative reasoning can actually get lost, because every once in a while, I’ll explain the idea of relative risk to a statistic student, and they don’t get it. They didn’t know about it beforehand. It’s the most basic stuff that I think we should be teaching everyone, that I think, actually, it’s most important that our statistics students take this. Almost a quantitative reasoning, a critical thinking course on life, I think, really, should be implemented in every college across America.
Bailer: Well, I feel obligated now that I asked the question in the other direction. And I think you’ve actually have touched on it a bit by your advocacy of critical thinking and quantitative reasoning about what should a journalist take, but more broadly, what’s your perspective on people that are trying to get a good, solid foundation in data literacy and statistical literacy. What might be part of that story? What might be part of that preparation in your mind?
Vittert: I think there’s a couple programs that have tried to do it, but it’s this – you don’t need to know how to do a t-test in order to understand critical thinking or quantitative reasoning of numbers. You need to understand that sometimes there’s a plus or minus gain. You need to understand that sometimes results happen just cause, and it doesn’t mean that one thing’s better than another. You really need to understand this whole idea of where numbers come from, and I really think we miss that a lot. We miss that just basic understanding. It’s that same idea that if you’re teaching a kid to read, and the kid can’t learn to read, no one ever says, oh, it’s OK, he doesn’t need to learn to read. I didn’t learn to read, and I did fine in life. We never hear that. On the other hand, we’ll hear parents say all the time to their kids, oh it’s OK, Johnny or Susie. You don’t need to learn math, because I was really bad at it, and I did fine in life. So, I think it actually goes back even further to when we’re kids, and to say, everyone can learn math, we just need to find different ways to teach it to them.
Pennington: I was going to say, because I was reading – I cannot remember where I was reading this – where you had a math teacher tell your parents that you were no good at math, and they should just give up on you learning math. Or is that – am I getting that right? Wasn’t that something you talk about?
Vittert: Yes, when I was 14, I was told – I had failed a, I think it was algebra or geometry class – and the teacher just said, she’s terrible at math, she’s never going to be able to go anywhere with it, and I’m not even sure she should be at this school.
Bailer: Oh, how affirming.
Vittert: Yeah, made me feel really good, but you know, my mother said, I don’t care, I want her to learn it, I know she can do it. And I was lucky enough to have a mother that really believed that. And she got me a tutor, and I was lucky enough to be able to have a tutor, and they just helped me think about it differently, and see the numbers in a different way. And I think it’s the same thing, not everyone learns to read the same way. There’re many different ways we teach kids to read, but there’s only one way we teach kids math, and I think that’s the problem.
Campbell: Why do you think numbers are so scary to so many young people? I mean –
Pennington: They’re scary to me sometimes, Richard.
Campbell: I know, well, to me too. But where did that come from? I remember being pretty good at math in high school, but at some point, I lost that, and I don’t know why.
Vittert: I think – we hear all the time that it’s scary. It’s that reinforced thinking. People – it would be embarrassing for someone to say, I don’t know how to read. It’s not embarrassing for someone to say, I don’t know how to do math.
Campbell: Yeah, some people are proud of that.
Bailer: Yeah, it’s a badge of honor.
Vittert: Right, it’s a badge of honor, and I think that’s the real problem here. It shouldn’t be a badge of honor. And I think it’s our duty to figure out new ways to teach people to learn, so that they really have to learn, and I think it’s really important for parents, even if they did hate math, to not tell their kid that. It’s not an affirming thing, it’s a, OK, how can we fix this? How can we make you better at it? How can we figure out new ways to learn?
Pennington: I’m going to go back to that question I was asking about, sort of, your experience as a teenager, because you were also – I think one of the BBC’s 20 Ambassadors – or yeah, for women science experts or something like that. And statistics and fields that seeing math, hard sciences, these are fields where it’s very difficult to be a woman at times, and part of that comes from being a young kid, and I had a similar experience to you, where I had a math teacher who didn’t tell me I couldn’t do it, but just basically made me feel completely dumb, and I think that’s partly where my hate of math came from for a long time. And then I got in to grad school and I’m like, I love stats actually; I wish I’d known this younger. What do we do to create a climate in which young women feel like stats, or math, or whatever sort of science-y, math-y thing they want to go in to, is a safe space for them to go into, and something that they can pursue, and they’re going to be supported?
Vittert: I think it really starts with parents. I wish I could say we can swoop in and fix everything, but it really starts at home with parents not saying, I hate math; you can’t do math. I think that’s – the confidence is built, at least in my personal opinion for kids at home. And further to that though, we can give them strong female role models. If you go to college – I have to say, in college, I didn’t have a single female professor. Not one. And I had wonderful male professors, and I’ve always had spectacular experiences, but I was not taught by a single female professor. And that’s hard, if you never see someone who’s a female professor teaching you, you’re going to think, well, maybe I shouldn’t do this. So, it’s not that the male professors are bad, they were some of the greatest people – I have some of the greatest relationships possible with my male professors, but I think it’s important that girls see females doing this, so that they know that they can do it.
Bailer: We host these careers involving quantitative skills for high school women every January on campus, and one of the things that we do is, we have women faculty and also graduate students that are leading activities. And one of the things that we’ve found is that the high school women that come up are just really delighted and engaged. And we found that it’s attracted a number of students to participate and come to the university. So, I think that those types of opportunities that create those, can really add value and really engage.
Vittert: Absolutely, I completely agree with you.
Pennington: So, what’s next for you? You’ve been appearing on Scottish news programs, sort of, breaking down stats information, you have a cooking program, you have a lot of things in the hopper. Do you have any new media on the horizon? What’s next for you, as far as, sort of, this kind of stuff goes?
Vittert: Well, I’m really excited that next year I’ll be a visiting assistant professor at Harvard with Xiao-Li Meng, in the Statistics Department. So, I’m really excited about that, and it’s with the launch of this Harvard Data Science Review, which I think will be an incredible opportunity to showcase statistics while being a scholarly journal, but in a whole new light of helping everyone understand this. So, it’s not a scholarly journal that only statisticians are able to read. And one of the things I’ve been working on a lot recently is writing, and I’ve loved doing it, and I’ve been really lucky to have some wonderful editors, which is the key, I’ve learned, to writing, is having someone else read your work that really knows what they’re doing. So, I think more on this, sort of, how do we really communicate numbers to people to the public, and how do we really empower the public to feel like they can read a news story and know what’s going on and know what questions to ask about the data and about the numbers.
Bailer: Just as a quick follow-up, so, this discussion of data science review and that effort, is really intriguing and we’re going to have Xiao-Li as a guest later on in Stats and Stories, but I’m curious, what is your perspective, and how do you differentiate between data science and statistics?
Vittert: That’s the thing I’ve been asking myself, because I’ll go to a dinner party, and if I say I’m a statistician, there’s like empty seats on either side of me [LAUGHTER], but if I say I’m a data scientist, all of the sudden I’ve become much more interesting, and I haven’t quite figured out what the difference is between the two. So, I think it’s a really good question. I think it’s something that the statistics community is struggling with a little bit of where’s the identity between statistics and data science. And like in any profession, you have different people feeling like, oh, well, if you’re a statistician, that means you do methodological research, or that means you do health statistics, or – no one really knows. So, I think one of the fabulous things about establishing this Harvard Data Science Review, is really, some time and that’s really what Xiao-Li’s been doing is, how do we really define this discipline of data science? What does it really mean to be a data scientist versus a statistician, and I really don’t think that question has been answered yet? And I think that’s what’s going to be so great about this is, trying to define it as a discipline.
Campbell: So, do you feel any obligation or special responsibility – this is sort of a big picture question about the, sort of, anti-evidence, anti-science, the idea that, whatever you believe is true. Now this may be connected to the whole number’s thing, but certainly, in the United States, I think this sort of undercurrent of dismissing data, dismissing evidence, is troubling, I think, especially for those of us who are teachers.
Vittert: Absolutely. I think it’s a really scary time that people are dismissing evidence. On the other hand, I do think we – and I include myself in that – are partly responsible. We have spent – when we publish this stuff, or we have stories, or press releases are sent out from our research, that are really misleading. You can’t expect a journalist to also be a statistician. So, when some of these press releases are sent out, or the research is done, and the study, or whatever, I think we actually are quite culpable in a lot of our work in terms of having the public be completely confused by what’s going on. You know when statistics are put out there, you know, one day a glass of wine’s good for you and the next it’s terrible for you, what do you sort of expect the public to do? I mean, they’re going to get really confused, and they’re going to start to dismiss the information completely. With the p-value crisis and all of this stuff going on, I think at some point, we really have to take responsibility as a profession to say we’re really – I hate to say it like this, but we’re going to sort of police the data much more strongly than we are now.
Bailer: I’m going to push back a little bit on that, just with the idea that, some of the issue of headlines and the research that you’re describing, is not necessarily something that a statistician would have control on, Liberty. A lot of times, that’s going to be embargoed by journals and then pushed out, because of the interest in the headline it generates. And so, I don’t know that there is necessarily a way for the stat community to help with that filtering. I mean, we can advocate, and we can certainly do what we can. I think that some of the leadership of organizations like the American Statistical Association, to jump into the p-value’s concern of malpractice, is an example of the stat world trying to weigh in, but when you think about things like your bacon story, or your French fry story, the stat part was probably a pretty small part of what lead to that being pushed out, don’t you think?
Vittert: I completely agree with you, and I think a lot of the work that’s already being done, is incredible, and I think we just need to do more of it. I do think that there are times where we – but that includes that, that includes us saying, wait a minute, this is ridiculous. It includes more people coming out and saying, this is ridiculous; we can’t do this. And it comes in to saying that, when you’re writing a paper with, I don’t know, a doctor or whoever, saying that I want something at the bottom of that paper that says, this is explaining what the significance means, and explaining where these things come from. But absolutely, I think the ASA and the RSS are doing everything they know how to. I just think that we need to keep doing more, and we have to keep pushing forward, because I just don’t see who else is going to do it if we don’t.
Bailer: Oh, yeah, Amen.
Pennington: That’s all the time we have for our conversation today, Liberty. Thank you so much for being here.
Vittert: Thank you.
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 Podcast, or other places you 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.