Calendar +

Stories, Weekly Interviews, things to see and footnotes.
Stories, Weekly Interviews, things to see and footnotes.
Stories, Weekly Interviews, things to see and footnotes.

Week 13 — July 27, 2018

In Conversation

Mona Chalabi Data Journalist11 minute read

Photography Emma-Lee Moss

Mona Chalabi is a data journalist. Having previously worked at FiveThirtyEight, The Bank of England, the Economist Intelligence Unit, Transparency International and the International Organization for Migration, she is now Data Editor at the Guardian US. An illustrator and presenter, she’s worked on shows for the BBC, Vice, and Channel 4. Here she talks politics, facts, the positives of Instagram, and why we should all be more interested in numbers.

Rosalind Jana

Interview Rosalind Jana

Photography Emma-Lee Moss

RHow did you get started in working with data? 

MI actually started out in the NGO sector working for an organisation called the International Organization for Migration. I was doing something called ‘monitoring evaluation’ for them, which is basically where you try to assess need in a country where there’s been some kind of humanitarian issue. That’s basically all about statistics: trying to measure how many people are in need, and then figuring out how much money you need.

But I got quite frustrated with it because a lot of the work we were doing was being scheduled by a small, elite group of people, and that makes it hard, sometimes, to spot errors… That’s a big part of my philosophy of data journalism: the more people you share something with, the more likely you are to know whether what you’re doing is actually correct and accurate and good. So, because the reports I was making were only being shared with five or ten people, it made me decide that I wanted to get into journalism.

Yes! I want to chat more about the relationship specifically between data and politics. I was watching your video on the dark power of polling, and obviously that’s a continually relevant conversation. I was wondering, could you tell me more about this? 

I think that polling has had quite a negative effect on the quality of our political debate. If you take the U.S. election, for example, people were so focused on who was going to win as opposed to any sort of meaningful debate. And even more problematically, I think you could argue that it even influenced some kind of voter behaviour. I spoke to people who said Bernie Sanders would be their first-choice candidate but they wouldn't vote for him because they knew he had no chance of winning — and the only way they knew that is because of polling. I think that’s really, really bad. I think you should vote for the candidate you want to win, rather than the one you think has a viable chance. And I also think the way that some of that information was communicated was slightly misleading. 

Again, you think about the web page of a website like Fivethirtyeight which suggested that Hillary Clinton was going to win — sorry, I’m really getting into the weeds already here! — and it communicated those probabilities with decimal places, and those numbers were accurate [enough] to have decimal places. This is a big thing that I focus on. You shouldn’t put decimal places on things that are fundamentally inaccurate. So yes, I’m very, very sceptical of polling in general — not only its accuracy, but also its usefulness, basically. 

Off the back of that, given that words like ‘truth’ and ‘facts’ and ‘polling’ are currently at the forefront of a lot of dialogue — beyond registering your scepticism in a public forum, how do you view your role in those conversations? Do you feel any responsibilities towards certain types of reporting? 

Yeah, I think there’s definitely still a role for data to play. The fact that I don’t really trust polling doesn’t mean I think numbers are useless in political journalism. I think there’s a really important role for national statistics in communicating the political record of certain figures or political parties. 

Also, for a while now I’ve been writing a fact-checking column that uses data — but the idea of that column is that, you know, if you’re feeling sceptical of the answers I’m giving you, I’m going to show you exactly how I got to those answers. Rather than just saying, “the answer is six”, I’m really going to show you my methodology for arriving at that answer, and I hope that’s going to help build some public trust around numbers and facts. 

Yes, that really struck me when I was reading and thinking about your work — because obviously my research is so different in its methodology. I guess I’m often diving into books or other written sources. Anyway, that’s tangential: it just hit me earlier, in terms of where we look and how transparent we’re being in flagging how we get to the conclusion of an argument. 

But even your style of journalism could easily adopt a lot of the same things. Let’s take this exact interview for example. Obviously, there will be some editing that goes into the piece before it’s up, but wouldn’t it be amazing if there was just a button and a link so that the people who were feeling sceptical about the way it was edited could listen to a full transcript and hear the intonations in voice, the hesitations… 

It’s a very true thing! Or hearing all the ‘likes’ or the pauses or the moments where the sentence goes halfway and then the thought changes direction. We’re all shaping stories all the time. I also want to talk about how I know that even I probably once fell into the trap of hearing a word like ‘data’ and thinking, “oh, math, that’s not for me.” But obviously your research has covered everything from the wage gap to masturbation to tattoos. Do you think there’s also something to be said for encouraging more people to think creatively about numbers and the impact they have on their real-life circumstances? 

Absolutely. I think a big part of my job is to demonstrate to people that these numbers represent you — whether you ask them to or not, they’re like these unelected political figures, if you like! So, you have an incentive to try and understand them. When there’s a figure that’s like, ‘two out of ten women hate men’, that figure represents you. There’s an interest in understanding that and [how it] relates very much to your everyday life.

Mona's Top 10 Books

New and Old Favourites

I also love the way you use your sketches — both on your Instagram and elsewhere — to contextualise things. What are the practicalities of that? How do you go about distilling down all that information in a snappy visual? 

My process totally depends on how much time I have. So, the first step is always to find the numbers, and find the most accurate numbers I can. And then there’s a process of synthesis. The next step is trying to figure out the… way to draw it. Sometimes there’s more of an emphasis on doing something that looks fun. Sometimes there’s more of an emphasis on doing something simple. And you always take into account subject matter. For example, I did one about the size of solitary confinement cells, and in that particular instance, I think it wouldn’t be comfortable to choose a tone that’s… something emotive or funny, a prisoner sat there, so I went for something really simple. 

Ideally, when I have more time as well, I think it’s really, really important to run the draft graphics I’ve done by someone who either isn’t interested in data or doesn't have the expertise… just to make sure it makes sense. And that’s why I think Instagram is a really helpful platform for me, because I’ve posted up so many where in the comments below people [will say], “I don’t get it.” Or people who have more expertise on that particular subject matter call me out and say, “you misunderstood this.” For example, I remember after Muhammad Ali’s death, I put up this data visualisation about how much of a champion he was and how many fights he’d won, and I just misinterpreted the data. The data set was actually how successful each of his opponents had been before facing him. And people who knew more about boxing than me wrote in the comments, “oh you got this wrong” — and it’s so helpful. It just means I take down the data sketch, and I respond, and I keep that in mind for the future.

That’s a really great thing to hear too, because I feel like a lot of the conversations we have about Instagram aren’t about the positives and the possibilities that come with having the platform for instant feedback. That’s not a conversation I’d really thought of.

I honestly find the quality of comments around data pieces is so different to ones around comment pieces. Right from the beginning when I started at The Guardian, I remember all of the comment pieces had to be massively moderated — so many comments were deleted. And when you publish a piece that focuses on numbers, people comment saying, “I think the spike in this year can be explained by this…”, or, “here’s a better data source.” I feel like it really elevates the quality of debate. It’s so helpful. 

Talking of interpretation and the visual, I was looking at your exhibition a few years ago, Photographs by Numbers. In an accompanying interview you said, “art is about interpretation anyway and statistics are about interpretation.” 

Oh my god, I have no recollection of saying that ever! [laughs]

That’s also the problem with interviews — when people pull up a quote, and you’re like, “Did I say that? Really?” 

I’m happy. That sounds pretty smart for me. I’ll go with that. 

Well I loved it, because it really made me think about the limitations of both — especially in how we form narratives about places or reduce people down to a set of statistics. I was wondering if you could expand on that idea? 

I choose to do hand drawn [data visualisation] because when you look at an illustration it’s impossible to forget that a human was involved in its creation. Whereas when you look at a computer generated graphic, sometimes you forget that a human played a role: a human with their own biases, their own interpretations… At every single stage of the data collection you’re making so many choices that are about you as an individual, that informs how you interpret the data. So, it’s really important that people don’t lose sight of that. I think it’s one of the reasons why people are really sceptical of numbers today, because they instinctively know that there are those biases. So many people like me: we’ve such an incentive to overstate accuracy. We claim this is beautiful, objective science, but it’s not. It’s important that we’re more honest about that. 

Absolutely. To skip to your column Dear Mona — which I loved!  were there any particular favourite questions you investigated? 

There were lots that were not published. Someone wrote to me to ask if I was wearing socks, and I replied being like, “the statistics on sock wearing?” and they were like, “no, I just want to know if you’re wearing socks right now.” I think someone once wrote to me to ask the probability of a bee flying into a bumhole. 

Wow!

I did a talk where I actually went through all the funny questions I got over the years — screen grabs with the names cropped out, obviously. Anyway, I was going through all these questions and I came to one that was, “How much pee is a lot of pee?” And lots of people after the talk were like, “No, you actually need to answer that question” — so that was another interesting exercise in ways to communicate data. I thought about volumes and I thought about how 150ml is… tricky to imagine. So I thought of drink containers, which I know is kind of gross. I used different drink containers to show what was normal urination volume and what was abnormal in terms of too little, and too much. That was one of my favourite questions. 

In your closing letter to that column you said, “You showed me how data is imperfect and your personal experiences helped me fill in the “whys” that are so often missing in my work”. I was wondering, do you ever apply data to your own personal experiences? 

That’s so interesting. I can give one really great example of the why. I once had someone asking me, “Are people in prison less likely to be atheist?” So I wrote up this piece and looked at different data groups of prisoners, and you see that Muslims and Jewish people are over-represented. And I said, “You know, we’re not quite sure why this is, we don’t know if they’re converting in prison, blah blah blah” and, again, the readers really helped me to understand the why. I got emails from former inmates saying that part of the reason for conversion, if you take the more cynical view, is that if you convert to Islam you get more time out of your cells to pray, if you convert to Judaism you get kosher meals which are better than the normal prison meals – and there’s no way I would have understood any of those nuances by just looking at the numbers in a spreadsheet. 

I’d say that maybe numbers have helped me in my life. With the Vagina Dispatches, I was looking at statistics on women who get thrush or infections as a result of using body wash on their vaginas. I would absolutely never, ever use any kind of product to wash my vagina now – never, ever, ever. And that’s based on numbers around that stuff. I think if I ever wanted to have kids, I would pay pretty close attention to the probabilities of conceiving at any given age to help me figure out when I want to do that. Maybe that’s a weird example. 

Not at all – especially when you have the knowledge and the skill set to maybe think about it in terms that other people can’t. 

But I think it’s so helpful – because I hear so many women in their early 30s freaking out. But actually, they have a little more time than they might think they do. To know that you can continue on with your career for an extra one or two years could make a huge difference to your earnings, to your life, to your happiness… 

Talking of Vagina Dispatches, you’ve kind of already answered this – but I was wondering if there was anything in particular you took away from working on it as a full series? 

Oh, so much. It made me realise how ignorant I was previously about a body part that I was carrying around 24/7, and that I thought I knew a decent amount about. Also, it just informs a lot of my journalistic practices – because I feel like part of the reason why people reacted so positively to it is, again, that it wasn’t this forced objectivity. I wasn’t just this person handing the microphone over to people and getting them to share their experiences and wrapping up at the end. It was very obvious that every single subject affected both Mae [her co-creator] and I directly, as well as the people we were speaking to, and I think that journalists always strive to be honest. That’s part of being honest: staying upfront in the ways that a subject does or doesn’t affect you, too.    

Finally, are there any particular ways you’d like conversations around female sexuality to advance or be tackled further, based on what you covered in the show? 

I, personally, would love to see more research done on how female sex drive compares to male sex drive. I think that’s really fascinating, having spoken to a lot of female friends of mine who are in heterosexual relationships and just find a lot of the social assumptions really frustrating. Actually, [what] I’m most interested in is health things… The safety of drugs and other things for women’s bodies. I was reading something yesterday on how a lot of condoms contain petroleum, which is massively damaging to women’s vaginas. You know, all these products! Loads of lube is really, really bad… That’s so messed up that we are damaging our bodies and none of these manufacturers really, necessarily care. And they partly don’t care because we’re so ill informed that we’re not putting pressure on them to come up with new products.