Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #37: Sarah Kliff
136 Plays9 years ago
On this week’s episode of The PolicyViz Podcast, I chat with Sarah Kliff from Vox. Sarah is a Senior Editor at Vox.com and oversees reporting on health and medicine. She previously worked at the Washington Post and Politico. Sarah and I...
The post Episode #37: Sarah Kliff appeared first on PolicyViz.
Recommended
Transcript
Introduction and Sponsorship
00:00:00
Speaker
This episode of the PolicyViz podcast is brought to you by Juice Analytics. Juice is the company behind Juicebox, a new kind of platform for presenting data. It's a platform designed to deliver easy-to-read, interactive data applications and dashboards. Juicebox turns your valuable analyses into a story for everyday decision makers. For more information on Juicebox or to schedule a demo, visit juiceanalytics.com.
Guest Introduction: Sarah Cliff
00:00:35
Speaker
Welcome back to the Policy Viz podcast. I'm your host, John Schwabisch. I'm joined today by Sarah Cliff, who is the senior editor at Vox.com. Sarah works on a variety of issues, mostly health and medicine. Prior to that, she was at the Washington Post and Politico. So, Sarah, welcome to the show.
00:00:52
Speaker
Yeah, thank you for having me. Thanks for coming on. I'm very excited because I've been following a lot of the things you've been writing about lately.
Discussion on Trump's Healthcare Plan
00:00:59
Speaker
Of course, today being March 3rd with Donald Trump's new healthcare plan just released, you had a big article about that. And so I want to talk about a few of the projects that you've done at Vox, but let me start by asking how does Vox differ from some of the previous places that you've worked, like the Washington Post and other places that are sort of these, you might consider them more traditional news agencies.
00:01:17
Speaker
Yes, I'd say there's two big differences that stand out to me. I had my two-year anniversary last week. Congratulations. Yeah, thank you. I didn't get any gift. I don't know what the traditional two-year gift is because I have to ask about that. Anyways, the two big differences that I notice, one is just not having a fringe component at all. Everywhere else, I've worked up until now.
00:01:40
Speaker
has had at least some sort of print edition. In Politico, it wasn't every day. It fluctuated with whether or not Congress was in session. But you always have to think about what things would look like in print. At Fox, we don't have that restraint. So we're basically only working on one medium. We're working online. And that allows you to focus your efforts a little bit on one type of reader and think about how they would want information presented.
00:02:06
Speaker
So it's almost a little bit freeing in a way. We can talk later because there's so many different ways to present things online. Like you can do a chart, you can do an interactive chart, you could do a GIF of a chart. So there's a lot of options, but you're working in one medium essentially.
Vox's Approach to Digital Content
00:02:19
Speaker
So it's one big difference. The second one I'd say is since we're a pretty new outlet, there's no one to say, oh, that's not the way we do things here because we literally have no way that we do things. So there's like a lot of experimentation right now. We're trying to figure out, you know, how people,
00:02:36
Speaker
interact with our website, how they interact with particular graphics. One of the examples we're seeing right now is that we really like building these data-rich interactives that people can explore and poke around with.
00:02:50
Speaker
I think that's so cool we're giving people access to all this data. It turns out no one actually clicks on that. Sometimes they just want a flat chart and they just want to look at it and load faster on their phone. There's a lot of experimentation right now about what is the best way to present information and it's especially easy to do in an environment where you have very few things figured out yet.
Adapting Content for Different Platforms
00:03:14
Speaker
And do you find that you can easily, or maybe there's a categorization of the types of users who want the static chart versus the in-depth data product that they can play around with? Yeah, I think it's not the type of person, but more of where they're reading. So if you think of you're on your desktop and you're reading a story, it's pretty easy to explore and click around and change things in the story. But if you're on your phone, I think what's most important to you is a really fast experience that you just
00:03:45
Speaker
You don't want to get bogged down with this interactive that might be hard to manipulate on a small screen. I think it's mostly about where people are reading and less about who they are. I'm curious, do you find that when you're working on an authority that the content is driving how you're going to present the information or more about thinking about who the person is that's going to be reading it?
00:04:08
Speaker
Yeah, it's definitely both. So I think content is probably first where we're thinking about like, if we have something that'd be interesting to show in a visual way, like what is the right way to do that. And then you know, one of the things we've had to think about more and more because we know that I believe the latest numbers are about 45%.
00:04:26
Speaker
of Vox.com readers or on mobile that we have to think through, like if we want to build this graphic, will it work on a phone? So we kind of start with like what we want and then think through like how is this going to work on the different platforms that people are people are reading us.
Collaboration with Researchers
00:04:40
Speaker
I'm also curious, a lot of your writing seems to be based on research, academic research, on health, on healthcare, on health policy. How do you work with researchers to sort of get that information out of the academic journals, out of the working papers and onto Vox? And do you think there's a place where researchers should be thinking more Vox-like is for lack of a better word. Do you think about trying to communicate their work to a broader audience?
00:05:04
Speaker
Yeah, I think there's definitely space for that. So the way I work with this is there's a handful of academic journals, mostly in the health policy space that I read pretty regularly. This would be things like Health Affairs, the New England Journal of Medicine, the Journal of JAMA, the Journal of the American Medical Association, think tanks, liquors, you know, those sort of
00:05:25
Speaker
Things I'm generally, because I've been covering this for about seven years now, I've generally gotten to know a lot of people in that space and give a good, pretty good sense of the landscape. So it's always helpful to me when researchers reach out to me and let me know like, Hey, we have this thing we're working on.
00:05:41
Speaker
that's going to be coming out. The IUD study you mentioned that, no, you didn't mention that. We mentioned that before we started taking it, so nobody knows about that. I wrote a story about the unplanned pregnancy rate being at a 30-year low, and that IUDs are a large part of that. That was a story where a day before the research was coming out, the Guttmacher Institute, which had two authors on the paper, reached out to me and was like, hey, there's this paper coming out. Let us know.
00:06:10
Speaker
if you want to talk to our researchers. So I think reaching out to the media is always helpful in terms of getting your research out. And then can I actually, I have one specific plea to any researchers who happen to be listening to this. One of the things that's a little hard about covering academic studies that I run into all the time is a lot of times, I don't know if you could shed some light on this, a lot of times there's charts, they show some kind of interesting trend and like, oh, that's so cool, I want to grab the numbers for certain years
00:06:39
Speaker
And then you can't find the data. And it's like this constant frustration of mine reading papers where I want to write about the data that's being expressed somewhere, but the actual numbers aren't available. So that's my one plea to researchers. Open your data. Yes, open your data. And I don't think anyone's doing it out of ill will. It's maybe an oversight.
00:07:01
Speaker
But if you're looking to fully share your research, making all the numbers that you're displaying in a graphic way also available as numbers is an awesome thing in my book. Yeah. I mean, I would second that. I think there's a couple of reasons why.
00:07:16
Speaker
researchers probably don't release the data along with the reports, although now I'm seeing more academic journals require authors to publish the data along with the article. Oh, interesting. So for the American Economic Review and the Journal of Economic Perspectives, you have to put your data up as well. Oh, that's great. But I think there's a couple parts to it. One is that it's not in the DNA of researchers to sort of like open everything up.
00:07:39
Speaker
And I think another reason is people feel very protective of their work, that this is a product that they've been doing for a long time and it's their project and that's the way it's going to be. What often happens as an author, you try to put everything you can onto the figures or into the article. So you pack a scatter plot, for example, with every label of 75 points and someone gets anything.
00:08:02
Speaker
Whereas if you said here's just a couple labels and the excel file is CSV or whatever is on the website then you sort of free yourself up so infuriating things I can see they have the data and they're willing to like sort of show it to me but not give me the not the whole thing yeah that's the hard thing I run into yeah and like to be like this is true of the.
00:08:21
Speaker
I'm playing pregnancy article I wrote where it had a graph but not the numbers used to make the graph so I emailed the researchers and they very kindly sent me an excel sheet right away without the information right but it'd be so much easier if it was just it was just there right there yeah yeah do
Making Content Shareable on Social Media
00:08:36
Speaker
you think
00:08:36
Speaker
that researchers aside from opening their data, do you think they can learn how to communicate in these different ways? Can they think about doing better with their graphs? Can you think about, I mean, I'm just looking at, for example, I'm just looking at the unplanned pregnancy piece that you have here. And the title on the first graph is unintended pregnancies are at an all-time low.
00:08:55
Speaker
Now, my guess is the report has some sort of like nondescript title, but are these the sorts of things that researchers should be trying or should they try to stay in their lane and open their data so that people who are like you writing should take the data and do more with it?
00:09:09
Speaker
I don't know if researchers would feel uncomfortable with it, but I think if you're trying to get more interest and eyes on the things you're working on, one of the things, the headlines on our graphs are pretty intentional. This is something we've had to get better at is one of the things we've realized is that
00:09:26
Speaker
a lot of our graphics and visualizations, they don't just live on our stories. People screenshot them, people take them and use them other places. They go on Facebook, they go on Twitter. So we try and think about a lot of our graphics, like, you know, the one you just read the title of. I wanted to make sense in the story, but I also wanted to be able to live on its own. So like, if people want to share it on Twitter, it just like makes sense.
00:09:50
Speaker
On its own to the headline it's essentially thinking like every graph we publish on box has to have its own headline not just a story and that headline i think of it as a good entry point for a reader who's like scanning over this piece there like do i want to read this in the kind of big bold text i have i want to throw like a mini interesting facts that people to convince them that this is like something that they want to spend their time on when they have
00:10:17
Speaker
Like I'm competing against an entire internet of cat gifts or whatever. So I have to work really hard. So do all of us. So I think, I don't know how like academic journals or think tanks would react to it, but I think it's a good move and that anything you can do to kind of sell people. If you're doing this research, you clearly have decided like, this is interesting enough to put a lot of time into like, why not take the extra step of like,
00:10:42
Speaker
really using the tools you have to get people to look at it. Right. And do you find that writing in that way, writing sort of the more active headlines, but also thinking of a person may just look at the graph, they may just look at this part. Has that changed the way you approach writing a story?
00:10:58
Speaker
A little bit. I mean, one of the things it does for the better is it forces you to sharpen your point a little bit because you're kind of like, you're sometimes thinking about, well, if I had to like summarize this whole story in a tweet, like what would I be doing? And those constraints kind of, they make it harder to kind of wander around in your story and not quite be sure what the point is. And then, you know, you kind of know the point of every graph that you're including. Like I think I have another one lower down in there that's something like
00:11:28
Speaker
IOD use has tripled since 2007. And that's like another one where you have to think through like, why am I including this? What is this doing for me? And chances are like, if you can't write like a interesting headline for it, maybe it's not that useful.
Evolving Role of Data Journalists
00:11:42
Speaker
So I actually find headlines are hard to write. Doing it for every graph can feel like a little tedious and overwhelming. But I generally find like it leads to a better story because I'm kind of like thinking through like, well, what's the point of this being here? Right. Right.
00:11:55
Speaker
So there's obviously a lot of talk these days about data journalism. Vox came out around the same time as FiveThirtyEight at the time that people were sort of talking about this. Do you call yourself a data journalist?
00:12:10
Speaker
I don't. I mean, I wouldn't be offended if anyone did. I think you could actually read my body of work and be like, oh no, maybe she is a data journalist. I don't know. Most people here at Vox have made their own graphs, have worked with academic reports, have done a lot of things that might fall into the category of data journalist. This isn't just true of
00:12:32
Speaker
writers, but also, you know, people who make our videos and people who manage our Snapchat channel. Like all of them are working with numbers and things like one would consider data. You know, I've always just thought of myself as like a journalist and an editor. And like, I definitely write about data a lot. I'm not great at Excel. Maybe that's my like dividing line. So maybe the better first question is how would you define data journal?
00:12:59
Speaker
Yeah, that's a great question. I mean, if you kind of go broad and say like, well, you know, anything.
00:13:05
Speaker
involving data, then probably pretty much all of us are data journalists. Which seems odd because journalists have been using data forever. Right. And then it's not a new thing that Boxer 538 invented. It's this thing that's been going on that got a fancy new name. I think there's definitely an element of being able to work with big, complex data sets. And that's something I can't do as well as a lot of my colleagues here can. So being able, for example,
00:13:31
Speaker
like my colleague, who's a news app developer here, who was able to make her own database of every adverse reaction, anyone's reported to the FDA for a supplement, and then use that to understand which supplements, these nutritional supplements are most problematic. So I don't have the skill set to do things like that. So I think that might be one element of it that keeps me from considering myself a data journalist as this ability to work with.
00:13:59
Speaker
very large data sets. But the small ones, I'm pretty good with those. But do you think people who are new to the journalism field are going into school for journalism, do you think they need to be thinking of themselves that they have to become data journalists, that they need to be able to work with large data sets, that they need to have all these skills? I mean, do you think new journalists need to be sort of the unicorn? Or can they focus in and find something that hopefully that they really like, but they're also really good at and focus on that skill?
00:14:27
Speaker
Yeah, I think the answer is probably either. I mean, one of the nice things about the way newsrooms are going is it seems like there's a lot of different jobs that are available. We still have a lot of traditional writers and we have people making videos and graphics and people who are doing work with these larger data sets. So I think that group of people, just because data sets are getting larger and larger and there's more and more of them,
00:14:52
Speaker
that that group of people is growing within the newsroom so it's definitely a space to go into but i don't think it's a must have like when i look at the box newsroom and you know we're considered kind of a data journalism website we have maybe like two or three people with those kind of large database skills but then we have like tons of people who are very familiar
00:15:11
Speaker
with reading an academic paper, with making their own charts, with managing smaller data sets. So I'd say that's definitely a pretty widespread skill, but I think one that's pretty easy to learn more about.
Teen Behavior Project
00:15:22
Speaker
Right. So let's talk about the team then if you have a bunch of people coming together because you had this story that came out just a couple of weeks ago on today's team. So you want to talk about that story a little bit and talk about the team that you pulled together to put that out? Yeah. Yes, that's one of my favorite things we've done recently.
00:15:38
Speaker
And there's a project of myself and sue who i mentioned who's a news application developer here at box and i'm serif austin graphic supporter here the thing that made it work is super collaborative and basically how it started.
00:15:52
Speaker
was since I'm a healthcare reporter, I spent a lot of time with CDC data. And one of the things I noticed kind of like getting these every other year reports is that like teen behavior kept getting better and better, which is like something you don't expect. There's a lot of like, you know, I get off my lawn, like it looks like kids are probably up to no good sort of thing. But if you look at these data sets, and it's something it's you probably wouldn't notice that if you're just looking at like one report, like one annual release, but like since I've been
00:16:18
Speaker
Following this release, it's called the YRBS Youth Risk Behavior Surveillance Survey. Since I've been following it, I've kind of seen that there are these trends going on. And I kind of thought, wouldn't it be cool if we could show people how they stack up to today's teens? And that's where I knew I needed Sue's help because she can build something like that. So the story we built is one that changes depending on how old you are. So up near the top of the story, we ask folks to enter in their age.
00:16:45
Speaker
And then we can tell them, you know, teens today are X percent less likely to drink and X percent less likely to have sex. And when you were a kid, this many teens drink alcohol and now it's fallen to this number. So we were able to generate this story that was very specific.
00:17:02
Speaker
And the way we were able to do that is Sue and I literally sit next to each other at work. So I was constantly telling her, here's what my story looks like. And she's like, well, here's what I can actually do. Or, hey, I see that's an interesting data point. Here might be a good way to display it. So it's really, I think one of the problems that can often happen, I don't know if this is true in academia, but at least in journalism.
00:17:24
Speaker
a way a story like this might work is I might just send someone like Sue a complete finished draft and be like, here's my story, make some cool things for it. But this was much more collaborative. Sue was looking at what I was writing. I was understanding what technical things were possible. We were working in
00:17:42
Speaker
Tandem to kind of make the best project we can and turned out great. It did super well in traffic people really liked it and it was an interesting data set I think that's kind of where any good project has to start like the data actually told a really interesting story and that was really important to making it work.
00:18:00
Speaker
What I found really interesting and unique and great about it actually is that you have this opportunity as the reader to basically change the story, have the story become personalized for you. So you get right in the beginning, you have this little slider where you can select the year in which you as a reader were born, and then the text sort of updates with all the numbers relevant to your experience.
00:18:22
Speaker
And you don't really see that a ton in, even though we have all this data out there that people are telling stories about, they don't actually see a lot of this personalization of the data. Yeah. It was one thing we worried about because kind of to circle back what I said earlier is like one of the things we see in our analytics is that people don't like to click on things. Like you really, people just want to kind of use the swipe motion with their fingers and then go through stories. So if you're going to ask someone to do something, the payoff has to be good.
00:18:50
Speaker
And in this case, the payoff was good. Like you got a personalized story about that compared you to today's teenagers and people were interested enough that one of the things we saw, which is really surprising and rare for one of our stories.
00:19:02
Speaker
is that we had more interactions than we had page views, which suggests that people were going back and changing it because they wanted to see how the story changed when they changed the year. And that's very rare for our stories. Usually page views are much higher than interactions. So yeah, it was a lesson. We can get people to interact with their stories, but we also have to be cognizant of we're asking them to do something and to deliver a good payoff.
Impact of Personal Stories in Journalism
00:19:28
Speaker
I'm also curious and we'll sort of end here even though I think the next question we could probably talk about for a long time, which is on storytelling and stories more generally. There's obviously a big focus these days on telling stories with data or storytelling more generally. I'm not really sure how to define that to be honest. But this piece in particular on teens
00:19:46
Speaker
seems to get at more of the personal side. How important is it for you when you're trying to write a story to make those personal connections, either on the side where the user is going through the story or when you're interviewing people and trying to bring their story into an article where you're putting in a larger framework of data that you're using to tell a story? Generally, I don't know if I speak for Vox as a whole, but generally in my stories, I find that
00:20:14
Speaker
actual human stories are a very good way to get people interested in the data sets you're trying to talk about. A good example of that is I spent last year doing this project on fatal medical errors. The number of fatal medical errors, it's mind boggling and hard to wrap your head around. I forget the exact number, but I think most estimates peg it around 200,000 Americans.
00:20:39
Speaker
dying as a result of medical mistakes every year. And just kind of like this mind-boggling big number. And you know, I started off this project I was doing with an explainer with these numbers and like, it did fine. It got like some visitors. But the story that really hit was the story about this four-year-old girl in California who experienced a number of hospital acquired infections during her last year of life and died just before her fourth birthday. And that like really
00:21:06
Speaker
drove it home that these are not just numbers. There's a story about this girl who died a month before her fourth birthday because she kept getting these infections in her hospital and her condition kept deteriorating and there are her parents who have had to deal with the aftermath of this and the doctors who have had to deal with the aftermath of this. I think stories are
00:21:25
Speaker
very powerful. I think one of the hard things for us as journalists is often making sure the stories and the data line up. Obviously, you don't want to pick an outlier and you want to make the stories faithful to the data. Sometimes you have great data sets and it's really hard to find the right face for them. It's definitely a challenge, but I think it really enhances
00:21:46
Speaker
data actually having kind of people on faces and the people who make up the data associated with it. Yeah. This has been great. Thanks so much for coming on the show. Yeah, of course.
Promotion of Sarah's Podcast, The Weeds
00:21:56
Speaker
So everyone go out, take a look at Sarah's work on Vox. You also have a podcast, The Weeds, right? I do. Yes, The Weeds. Every week, usually on Fridays.
00:22:04
Speaker
usually on Fridays, so you should also listen to the weeds, talking about all of the interesting stuff that Vox writes about. Sarah, thanks again for coming on the show. This has been really interesting. And thanks to everyone for listening. If you have comments or questions, please do let me know. So until next time, this has been the Policy Vis Podcast.
Conclusion and Sponsor Message
00:22:33
Speaker
This episode of the PolicyViz podcast is brought to you by Juice Analytics. For 10 years, Juice has been helping clients like Aetna, the Virginia Chamber of Commerce, Notre Dame University, and US News and World Report create beautiful, easy to understand visualizations. Be sure to learn more about Juicebox, a new kind of platform for presenting data at juiceanalytics.com. And be sure to check out their book, Data Fluency, now available on Amazon.