Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #170: DataViz in the Time of COVID
185 Plays5 years ago
As the world hunkers down to slow the spread of the coronavirus, a lot of information is making its way to our information channels. Some of this information is true, some is false, and some is misleading in one way...
The post Episode #170: DataViz in the Time of COVID appeared first on PolicyViz.
Recommended
Transcript
Introduction to Responsible Data Visualization
00:00:11
Speaker
Welcome back to the Policy Viz Podcast. I'm your host, John Schwabish. And on this very special episode of the podcast, I sit down with four people to talk about the coronavirus, visualizing data about the virus, and how we can be more responsible in the way that we visualize and communicate our work.
00:00:27
Speaker
I'm very happy to have Alberto Cairo, Amanda McCulloch, Jan McNellow, and Ken Fields on the show coming from different fields of data visualization, information visualization, public health, and cartography to talk about all the things that they've seen in the last couple of weeks when it comes to visualizing data from COVID-19.
Ethical Responsibilities in Data Visualization
00:00:47
Speaker
It's a really important topic when we are thinking about visualizing and communicating information with an issue as important as the coronavirus pandemic because the data that we use and the visualizations that we create can affect people's lives and can affect communities. So I hope you will listen to this episode and take away some of the lessons learned that we've seen already in the last couple of weeks and you'll apply them to your own work.
00:01:13
Speaker
So this is a fairly long conversation, but one that I think is very important and very valuable, especially in this current time of quarantines and this pandemic virus. I'm also just letting you know that I'll be hosting a series of video, live video podcasts over the next few weeks. I'm already set up to have a number of great guests come on the show. On Thursday afternoon, Eastern time, I'll have Nigel Holmes join me.
00:01:42
Speaker
These will be video interviews. I'll be chatting with my guests for just a bit, and then I'll open it up for questions. So if you would like to ask Nigel Holmes questions about his work, or Harry Stevens will be on the show next week if you'd like him to ask about his work, or Anne Emery who will be on the show on Friday, please do come on. There is a blog post on policyvis.com that has all the links and the Zoom information that you'll need to join this conversation.
00:02:09
Speaker
So I hope you on your family and your friends and loved ones are all well, healthy and safe. And so here is my conversation with Alberta, Amanda, Jen and Ken.
00:02:24
Speaker
Hi, everybody. Thanks for joining me on a lovely blue sky Tuesday afternoon that I can see from inside my house. As I look around here, looks like everybody actually has sun, so that's nice. Thanks for joining me. Let's go around the room virtually, and then we can start talking about data visualization and communication about the coronavirus. Alberto, why don't you start us off?
00:02:48
Speaker
Sure. I am Alberto Cairo. I am a professor of visualization at the School of Communication at the University of Illinois. Hi, I'm Amanda McCulloch. I'm the data visualization lead at Excela. Prior to working there, I spent eight years working in global health and development and have my master's degree in public health. Hi, I'm Jed Manganello. I'm a professor at the University at Albany School of Public Health. I have a master's in public health and a PhD in public health, and I study health communication.
00:03:14
Speaker
Hi, I'm Ken Field. I'm a former academic from the UK. I now work at Esri in cloudy California today, and I'm basically a cartographer.
00:03:26
Speaker
Basically, basically. Well, thanks for coming together. It was great because we've got experts in all sorts of different areas, data visualization, cartography, public health, and I'm glad we can all spend a little bit of time chatting together. I wanted to start with a fairly broad and yet I think specific
00:03:45
Speaker
questions simultaneously, which is what are the considerations we should have as producers of data and information visualizations and data visualizations about the coronavirus.
Impact of Data Visualizations on Public Perception
00:03:57
Speaker
Amanda, you wrote this great post on Medium and you don't need to walk through all of your 10 points. I'll link to it on the show notes, but maybe you can encapsulate for us what are the key things that we as creators should consider. Thanks, John.
00:04:11
Speaker
So I think that we're in a unique environment right now with combating what is an emerging pandemic threat. We're still early in our understanding both of the virus and of all the data around the cases.
00:04:25
Speaker
around what we see from testing, around what we see in the demographics. And so I think the challenge we face is that we're in a really unique place. Unlike looking at disease burden data that comes from diseases that we've confatted and fought and have treatments for or have vaccines for, we're learning new things every single day. I find that I keep popping up. I see new preprints of articles popping up before they go through peer review just to try to accelerate the pace of information coming out from the research community.
00:04:53
Speaker
And I think as visualization designers, one of the exciting parts of being in data visualization is you can so quickly, with the tools available today, go from a structured data source to beautiful visualizations. And there isn't really any kind of vetting process for us to go from what we create on our computers, what we publish on our websites, share on Tableau Public or publish on Twitter. And so the challenge we face in this kind of emerging environment is that the pace of new information is fast.
00:05:22
Speaker
The pace of new data is fast. People get excited. But the charts and graphs that you create, well, it's always important not to mislead people. Right now, that can really come to some life and death harm because it can influence what people's individual decisions are around what actions they take. So I've seen charts that have persuaded people that this isn't something they should pay attention to because it's just not important, because it's not as bad as the flu. Those kind of assertions that come out of trying to benchmark
00:05:50
Speaker
early information about this disease against an existing disease that we know a lot more about. And so I think that as visualization designers, it's really on us to think meaningfully about what are we doing in terms of visualizing this data that's for our own understanding. That can be calming for some people, right? To dig into the data, to look at it in my own context, understand it from my own perspective.
00:06:12
Speaker
And I don't think anyone is saying, don't go play in and visualize the data for your own consumption. I think the big challenge is when we start publishing those visualizations into the public sphere. And in this kind of environment, I can't say, I can't count how many people have come to me and said, hey, can I ask you a question or hey, can I talk to you because you're a public health person and maybe you understand this and what should I do? And people trust the people who they know.
00:06:39
Speaker
And so if someone sees your really beautifully compelling graph that actually has a really overstated level of certainty around, say, the case fatality rate by country, which we still are trying to understand, and they get really either scared by it or believe that something isn't as necessary to act on, social distancing isn't as important.
00:07:00
Speaker
When they see that come from you, as someone they trust, they're going to trust you more than they're going to trust those kind of big news media sites in a lot of cases. And I've had these conversations with my parents back in a mid-sized Ms. Western town in Illinois, and about the information that they're seeing. And so for me, this becomes a really personal question of how are we as data visualizers amplifying and sharing the great work that's being done with data visualization in this context, flatten the curve,
00:07:30
Speaker
Planer Harry Stevens did, other information that's making up-to-date statistics available to people, and amplifying the great work that's happening. I don't think we need to each build our own new dashboard of case data. There'll be plenty of other data to analyze and visualize about this whole entire period around when all the social distancing is happening. I don't think that we all need to dive in and build our own dashboard of case data. Let's put our efforts and our energies to helping to support the public health organizations who
00:07:58
Speaker
might need some extra data expertise right now and be interested in collaborating. Let's put our time towards trying to understand the knockdown effects in the economy and with employment. But I think we have a certain ethical responsibility to amplify the things that help people understand what's happening and why social distancing is so important and to share that information with our friends and family and people who trust us.
00:08:21
Speaker
John, I think that that's the entire podcast. That's the end of the talk. I think that both the articles that I have seen written by Amanda these past few days really summarize all those points really, really well. And I think that anybody interested in all these issues should go ahead.
00:08:40
Speaker
and read those articles. I would just add myself that we could go a little bit more abstract in this discussion and we could use the lessons of coronavirus to think about visualization in general. I am right now working on a new book which is going to be published in the fall of 2021, which is all about how to make decisions about visualization, how to reason about visualization, how to make the right choices
00:09:08
Speaker
whenever we sign a visualization and
Assessing Credibility in Visualizations
00:09:10
Speaker
the first choice the first decision is should my visualization exist it's like should it be out there it's like and and there are different kinds of way to reply to this question so i believe that whenever you're designing anything that is related to an let's say inconsequential topic like you're going to design a visualization of
00:09:31
Speaker
you know i don't know a marvel movies or whatever that's have a lot of impact on you know people's lives so the question is why not why shouldn't i visualize this is so fun i'm going to go ahead and do it but when you're visualizing things that may have direct impact on people's lives and people's well-being is you should say should i really do this it's like
00:09:52
Speaker
is the opposite direction, is the opposite kind of reasoning. In one sense, the first case is, why not? And then the other sense is, should I? Those are the two questions that are complementary to each other, depending on the topic that you're going to visualize. Well, can I maybe just add there that I think you're hitting on an important point, which is that the data that we look at in public health, and especially in this kind of situation, all represent people. And I think sometimes we are taking ourselves away from
00:10:20
Speaker
these numbers all representing individual people who have names. And each of us can think about is, I think Richard Cogley put it well in an article she just posted today, that each of us has someone we can think of who's in one of these high-risk groups. And so as we're thinking about how we not just visualize data, but communicate about this entire ordeal, that we think about those people. And we think about how they would feel when we say things certain ways or use different kinds of language or seem to be the only impacting certain groups.
00:10:50
Speaker
and remembering that there are people behind each of those things. I just think this is so pointed. So one thing you mentioned, Amanda, was this idea of instead of worrying about creating things, we should be sharing things, sharing good things. And I know this is something that Alberto has written a lot about. And one thing to consider is, how do we assess whether something is of high quality come from a reputable source? And I know, Jen, you've done a lot of thinking on this. And so I'm curious about,
00:11:17
Speaker
how you think about when you look at a visualization, how do you assess the quality? In this case, I think the data is really the question, the most important question, is how do you assess the quality of the data before you start sharing it around to your networks?
00:11:31
Speaker
Yeah, I mean, that's such a great question. I think what everyone else has said so far, you know, everyone's making great points about what we need to think about going into the visualizations, but then we have to consider the other side of this and what's happening with the information that we're creating. So when I'm teaching my students or doing trainings out in the community,
00:11:49
Speaker
You know, I'm explaining to people, what should you look for in information before you believe it? You know, how do you assess whether it's credible and trustworthy and things like that? And, you know, think before you share, basically. And what's happening, what I've seen with a lot of these
00:12:04
Speaker
Visualizations is that there's a lack of information with them. So in other words, if I'm looking at a published journal article and I see a chart or a table there, then I know that, okay, that is coming from a journal article that's been published in an academic journal. It's gone through peer review and I know I can trust that information.
00:12:24
Speaker
A lot of the charts and graphs I'm seeing going all over social media right now, there's no clear identification where that's coming from. So a lot of them sometimes there's no text on
Accessibility in Data Visualizations
00:12:35
Speaker
it. There may be something about where the data came from, but it doesn't say who created the information, who created the actual graphic.
00:12:43
Speaker
It doesn't say the date. So for all I know, the numbers are from a month ago and not from today. And so what I really am having a hard time with is I'm getting a lot of questions from people who know I'm in public health and they're saying, well, what should I do? Is this correct? Is this information accurate?
00:13:00
Speaker
And I'm looking at the sources they're showing me and I'm not even able to assess, you know, how credible or valid they are. And so I would just put out a call to everyone when you're creating these great data visualization pieces to include that kind of identifying information and even a link to the original source because that helps us assess the credibility, which is such a key component in terms of how we assess the information and then go ahead and share it.
00:13:25
Speaker
Yeah, maybe the best example of that is the original flatten the curve diagram. We could argue about whether it's a data viz or it's a diagram. I think that's maybe later. But I think the first person who sort of circulated that on Twitter that where it got picked up, like no one really knew who had created.
00:13:42
Speaker
That's right. And so then first of all, the person who created it is not getting the credit for it, right? And then there were people making tweaks to it, and they're not sure, well, who do I talk to? And how do I let them know that I want to change this or whatever? So that's really critical. And so these things can go and get shared thousands and thousands of times, and then no one knows where the original author was to go back to.
00:14:03
Speaker
So that is really important. And I think that thinking about some of these best practices in terms of how we include identifying information on the visualizations is so important. Even if you're putting it in a blog post or a news article or something like that, people are taking screenshots. So if you don't have that identifying information right on the figure itself, it can get shared and no one knows where it came from, even though you might have written a great article explaining the context for that piece.
00:14:31
Speaker
The other thing I just want to mention is we have to remember too that there are people out there who may have disabilities who need some, you know, text descriptions in terms of going with an image and things like that. We want to be really in tune with that so that everyone's able to access the information and we're not creating any kind of a knowledge gap or disparities in terms of what people can access.
00:14:52
Speaker
And the other thing is a lot of people don't have computers now. They're just accessing these things on their phones. And so we want to make sure also that they can be displayed easily on mobile phones and that that identifying information can be seen on the phones as well as when you have it up on a full screen computer. Yeah.
00:15:10
Speaker
So I think there are, by my count, I think there's three visualizations that have sort of made their way into the common lexicon right now. There's a flatten the curve graph. There is Harry Stevens' interactive animated piece from the Washington Post, and I think
00:15:26
Speaker
more or less John Byrne Murdoch's line charts that are showing the curve flattening, right? And a lot of people have done a version of that, but I think John, the Financial Times are the ones that are sort of keeping that one up. And then there's a whole other class, I think, that are various types of maps, which is why I wanted to get Ken on the show to talk about what he's seen. Ken, you had very early on, very early on being like,
00:15:50
Speaker
10 days ago. You had this great Twitter thread about what to and what not to do with maps and the virus. I'd like to just ask you to maybe chime in here and talk about what you're seeing when it comes to making good and bad maps in this time. Sure. I think generally speaking, and everybody said the same kind of points, what we're really seeing is
00:16:15
Speaker
an amplification of just the general problems of production and consumption of information and the way in which the internet drives that. So yeah, it used to be the case when you're an academic, you could wait a year and a half to get your paper out and then you'd read something that was peer reviewed and fairly accepted and repeatable and you believed it.
00:16:36
Speaker
You know, we perhaps sometimes used to buy a newspaper and, you know, it was based on editorial and reporting that had taken maybe several days or weeks to put together. And now we can download or scrape a dataset and have a map on the screen or a graph on the screen within about three minutes flat. And that's both good and bad. So, you know, we're seeing a lot of these dashboards and a lot of maps and a lot of graphs built by people who can access the data very rapidly. And that's terrific.
00:17:04
Speaker
but we're also seeing at the other end of the spectrum, a lot of other people just scraping the data and grabbing it and making maps and graphs. I have a fair idea who I would trust to believe the graph or the map in front of me, but 99% of the general population don't, and that's not their fault.
Media Responsibility and Public Trust
00:17:25
Speaker
They're busy being experts and everything they're experts in.
00:17:29
Speaker
I think it's beholden on all of us to actually just be responsible about what we do with data. And Alberto made the point, yes, sometimes we're dealing with very, very immaterial pieces of information. It's just fun. And other times we're dealing with some pretty serious data sets. And I think the more important the data set or the more crucial it is to public understanding of the situation and the circumstances, the more you really do have to think very carefully about what you're doing.
00:17:59
Speaker
So in terms of the maps I started seeing, I guess they fell into two categories. Of course, mainstream media outlets were always going to want to grab the data and push their view of the problem, obviously for driving clicks and views of their media site. That's their job, right? And there were very variable examples coming out of mainstream media.
00:18:25
Speaker
some typically excellent examples by the organisations you would expect. Washington Post, New York Times are doing some fantastic stuff and still are.
00:18:34
Speaker
And some organizations just seem to just throw the data at whatever wall was closest. And, oh, there's my map. Quick, put it on the website. And that's great if it wasn't for the fact that like a million people get their information from that website. And there's an asterisk there. I don't know if it's a million. I just made that up. There you go. And then you also get just random people on the internet putting stuff together, which can be really, really great. Sometimes that's where the insightful stuff comes from.
00:19:03
Speaker
A lot of the best maps in the world were never made by people with cartographic training. So, you know, they just hit upon a great idea with some great data at just the right time and it informed. There's also a lot of pretty poor mapping out there, which are making maybe basic technical mistakes with the map or the data, you know, maybe not scaling symbols correctly. Maybe they're using very alarmist colors. And these are all sort of things that could be very easily fixed.
00:19:34
Speaker
But the problem from the consumer's point of view is they don't necessarily know what they're looking at needs fixing. So they're just consuming the information, right? And their eyes, you know, our sort of perceptual and our cognitive systems are attuned to allow us to recover information in very simple ways, you know, larger symbols, bigger, you know, brighter colors. And this is fairly typical. So working with that rather than trying to
00:20:04
Speaker
Alarm people, I think, is a really important consideration with this particular data set. I mean, if I was going to map, like Alberto, the incidence of heavy metal bands in Europe, I'm going to go all in on crazy fonts and typefaces and all sorts of stuff, but this isn't really the time for that, I don't think.
00:20:24
Speaker
So let me play devil advocate for just a moment, although I agree with everything you everyone here said I think in the in the data is field at least we seem to feel like everybody making stuff is great because we sort of narrow and winnow down to best practices and things that are
00:20:44
Speaker
you know, it's actually creative and really helping in that way further the field of data visualization. And Ken, you just said there are people who are not necessarily experts in particular fields, but they end up making something amazing because they have the right data at the right time with the right form.
00:21:00
Speaker
So, and the answer obviously to my question is yes. The question of course is, do we need to sacrifice that debate and that discussion in the infobiz world for the accuracy of the data? But I'm just curious about how you think about balancing these two things. There's the data field itself, discussing and arguing and debating forms and function and creativity. And then there's the responsibility of the data.
00:21:24
Speaker
Due to the gravity of the situation, I am pretty extreme in my opinions. If you have nothing to add to the conversation and you have not consulted with experts, shut up. Don't publish anything. That's basically my take. I mean, it's a basic journalistic responsibility issue. It's like if you are going to publish something that may have impacts on people's lives in terms of how they are going to make decisions on how to protect their families and their friends and their relatives, et cetera,
00:21:53
Speaker
you need to be responsible. Just refrain from public. It's fine, as everybody has said, to design visualizations for your own consumption or to publish them in private forums, to discuss graphic forums. That's great. But don't put them in places where people can find them very, very easily, like on Twitter or on Facebook, unless, again, that you can really consult seriously with public health experts, with epidemiologists, et cetera, who can vet whatever it is that you're going to publish.
00:22:21
Speaker
And I say this because you know this, John, my own books. I give my own books and my own visualizations, everything that I do, to like a group of like 20, 25, 30 friends with expertise in different fields, economics, statistics, epidemiology, et cetera, et cetera, before I publish anything. And even if I do that, my books and my visualizations are sometimes are published with mistakes. Even if I do that, even if I am extremely careful,
00:22:49
Speaker
we don't make mistakes i still make mistakes so in a situation like these it's even a bigger problem so we need to be extremely careful. You're absolutely right alberto but the problem is that. I'll make up another start here eight out of ten people who are just messing around with data i'm gonna listen to that because they just gonna do it's a technical challenge.
00:23:09
Speaker
They're just doing it because they can and they've got a YouTube account or a Twitter account and they can push it out into the world. So that's always going to happen. I think from my perspective, the real challenge is how do you force the consumer to know to go to Alberto's graphs as opposed to
00:23:28
Speaker
just some random person's stuff. How do we know that you're the go-to guy for this? I mean, I think that, you know. Well, we don't. That's the problem. I'm sorry, I interrupted you, Jen. I think that it was Jen who was going to- Oh, no, go ahead, Alberta. I'll go next. Go ahead.
00:23:42
Speaker
No, I was just about to say that this is a, I mean, it's not a problem just with coronavirus. It's a problem with the current information environment, like anybody can publish. So it's something that I have been discussing, you know, in recent talks, and even how it just lie on us in the next book. Everybody's a publisher. Everybody's a journalist.
00:23:59
Speaker
So I don't think that there will be a short-term solution to the problems that these situations poses. There's a long-term solution, which is basically to essentially educate the entirety of the population on journalistic ethics.
00:24:15
Speaker
Basically, what you should do ideally as a journalist, you should double check, you should verify before you retweet anything, before you post anything socially and publicly, you need to verify things with at least two or three different sources that are independent from each other. And if you cannot do that, you don't put it out.
00:24:33
Speaker
That's basically journalistic ethics. It's an ethics that belong to a particular field, the field of journalism. Well, maybe we need to transform these journalistic ethics into universal ethics, into citizen ethics, into civics, right, at some point in the near future. I know that this is a long-term solution, but otherwise, what should you do? I mean, it's like, should we just, you know, I don't know, vet whatever is published on YouTube or beforehand? I mean, you cannot really do that, right? It's super, super hard.
00:25:01
Speaker
And I think you're both pointing out great issues where we don't have a certain element of control like we used to with the news media, with journalists, right? And there are people, for better or for worse, out there creating great information and not so great information. And so one of the things that I focus on in my work has to do with health literacy and e-health literacy.
00:25:22
Speaker
I've been creating different kinds of online tools to help people learn some of these skills they need to navigate the online health information environment. And so in other words, what can we do to teach people to identify what are the trademarks that they're looking for to know when something's coming from a credible source or to understand how it's been put together or how it's been created.
00:25:43
Speaker
and to really think about that and you know schools are starting to do that more and more. My son in middle school has had to take you know computer literacy and information literacy classes and so he's learning about a lot of these concepts but we still haven't always applied them well to health and so that's where a lot of my work comes in and so trying to also
00:26:04
Speaker
attack this from the consumer side and how do we build these skills and so even for myself one of the things I've been doing with my friends and family and on my own social media feeds has been to really help them to see okay here's some good sources of information here's a really solid article you can look at to get good information to try to help guide them to information sources and visualizations of data that I think are are right on spot on you know with experts and who are putting things out that have been vetted or
00:26:34
Speaker
verified by experts in the field. I've actually been sending a daily email to my parents and some friends with recommended reading. Neither of my parents is on social media. They get their local news from their local news station. They get their local news from their local newspaper, which I think they jokingly call a newsletter at this point.
00:26:56
Speaker
But they go ahead and that's what they get their news from so I've been sending them a roundup of recommended reading because I think that's important because they trust You to give them good information, right? I think the other piece that I want to come full circle on though is I don't want to I don't want to believe that we can't influence and teach and change our own profession I mean, I want to think that we can help people better understand the implications and the
00:27:22
Speaker
kind of ethical responsibility of what you're publishing in these kind of environments. I think training the consumers, building health literacy, helping people better understand how to consume information is also part of it.
00:27:33
Speaker
And clearly, it requires different information, right? The article John referenced at the beginning about 10 considerations of data as developers. That company reached out and said, we want to publish this but adapted for consumers, which is, at the end of the day, when I was revising it, a very different audience on how you're communicating and what your asks are. So how we communicate has to be tailored. We have to think about how can we
00:27:56
Speaker
how can we retrain or influence our own field, right? I mean, a lot of people on this call right now have written books about this topic, gone ahead and shared more information. And part of it, I think, is working with people who run some of the ongoing challenges like Makeover Mondays, tidy Tuesdays, whatever those pieces are, people who are putting data sets out in the world and having to influence where do you need to kind of lean out and say no to publishing a data source because
00:28:25
Speaker
the visualizations that could be created could have negative implications. I think of the data set from Makeover Monday all about suicide in the UK and the conversation that came out of that around the implications of publishing that data irresponsibly. And I think about the choices we make about what we, we give people and ask them to visualize and how do we partner those asks with expertise. If it's a training opportunity,
00:28:51
Speaker
then should those data sets be coming with an expert attached to them or someone advising on what those things are? I mean, I saw survey data that I worked with for years from the demographic and help surveys get butchered in the VIS5 International Women's Day makeover Monday kickoff because people didn't understand the underlying data that it was one data point per country that was specific to a certain set of already calculated percentages
00:29:19
Speaker
And they were summing them and averaging them and running different mathematics that you can functionally run the math in Tableau, but that doesn't mean we should. And to me, that's a lot of the same as we're seeing with the data, the things we're seeing with the case data are people saying, oh, just calculate a case fatality rate. That's easy. When really there are so many nuances to what that math really says and what the outputs really are and what math should we do and what math should we not do.
00:29:47
Speaker
with those data sets. I think that's the danger of some of these big data sets is they make it frustratingly easy to run calculations that are meaningless and shouldn't be run. One of the points that I made in how charts lie in the previous book is that arithmetic alone never gives you the truth, right? You need to have domain specific knowledge before you apply any sort of mathematical operation to a data set. And this is such a basic
00:30:12
Speaker
that is oppressive to me, that we have not embraced it yet. It's like how many states we will make if we don't have domain-specific knowledge? If we don't consult with the people who know how the data was generated, where it comes from, what the limitations of the data are, and so on and so forth, and we just mindlessly apply a mathematical function to that data, or a mathematical calculation, such as, again, estimating an average, or a medium, or whatever, right? Sometimes that is not meaningful.
00:30:40
Speaker
How do you know if it is meaningful or not? Well, you need the qualitative knowledge about the dataset, and only experts can provide that, fortunately or unfortunately. Again, as a journalist, this comes naturally to me, just because perhaps
00:30:55
Speaker
we were taught in journalism school, basically, you know nothing. It's like assume that you know nothing. You're always consulting with someone. You always need to assume that you are the conduit between experts and the public, and you are just a translator. In order to make this translation, you cannot do it alone. You should never do it alone, because if you do it alone, you are going to make mistakes. That's
Role of Domain Experts in Data Interpretation
00:31:20
Speaker
inevitable.
00:31:20
Speaker
So always rely on sources. That's why in journalism we put so much emphasis on always quoting where the information comes from or who are we consulting with and so on and so forth, right? So I don't know. I mean, I'm rumbling a little bit, but it's impressive to me that the visualization community still is basically having a discussion that should have happened years ago already.
00:31:44
Speaker
I just wanted to add to build on what Alberto said is even as an academic in public health, I mean, I have Stata. I know how to analyze data. I took biostatistics, but there's a lot of things that I would never think to do on my own. I would always go to a biostatistician first or an epidemiologist to say, okay, can you help me with analyzing this data or understanding what I should do with these regression models or whatever.
00:32:06
Speaker
I mean, I can do it. I have the skills and the tools to do it, but I just would never even imagine doing it. So it is funny to me that people all of a sudden have access to this data and just are like, oh yeah, this is great. I can just run whatever numbers I want. And you certainly can, but then it's true. You don't have the context to understand what you're doing.
00:32:26
Speaker
Or you create a forecast model or whatever, just because you have the data science skills to basically create the forecast model or whatever. It's like, don't do that. It's like, don't do that on your own. It's so basic. Just, you know, have some intellectual humility. That's the word. Intellectual humility is that you know what you know.
00:32:46
Speaker
I last because I said, Jen, I have the exact same thought is that I've been talking to my friends who are actual epidemiologists because I am not an epidemiologist. I worked in monitoring and evaluation and health information systems, which lets me have an understanding of the data, but I don't have that in depth expertise and then epidemiologists are saying well,
00:33:06
Speaker
I'm not an infectious disease epidemiologist. I really should be talking to the infectious disease epidemiologist instead. And I've seen that comparison with the like, you've got these people on Twitter who are like, I built a model. I'm a data scientist. I'm a data scientist. And then you have the epi community who's like, let us defer to the most expert experts. In response to my convert, my whole entire article, my favorite, my favorite response I thought that just exemplified this form
00:33:35
Speaker
was in response to sharing the article about considerations for visualizing this data, I said, as someone said, aren't you an expert in plots? Show us your charts. And I said, here are like four really great charts, most of which you had referenced earlier, John, and that you should look at and that help enable understanding this great Reuters story about the Korean story, visualizing, contact tracing. These are great. And he's like, I've seen those. I know about them. Show us your plots. And I was like,
00:34:03
Speaker
I just have been making this whole argument that we should be amplifying the people who are doing this so well and have the expertise and not trying to reinvent the wheel ourselves. I don't have something I need to create and put out in the world. I want people to look at the great work already happening. And then I want to borrow and enjoy the giggles of my one year old child in my limited free time I have when he's actually awake. So I'm just amazed by some of just the responses about
00:34:28
Speaker
But it almost seems like that's what validates your expertise in this field, is if you visualize something that it validates that you are expert in that area or that you are good at this. And I just don't think we need that. I think we need to amplify and share the great work that's already happening that is validated and has followed all those things that Alberto has talked about about the rules of an ethics of journalism. Ken, did you have something you wanted to add?
00:34:55
Speaker
I mean, I'm partly I guess, I guess there's the way the discussions go. And it's obvious, and I think we all know this, we all come from research based backgrounds. So we have that innate ability to question and to want to validate and then to ask more questions and to justify and to get evidence and to throw this around before we before we publish something, whether it's a blog or whatever. And so many people don't.
00:35:23
Speaker
and don't have that ability. In terms of trying to persuade people of a better way, I think that's really what we're all advocating as well. And that's certainly what I tried to do with the tweets and the blog from, what was it, two years ago? No, sorry, 10 days ago. And that's what I wanted to do. I really just wanted to throw out a really quick, easy, here's a few things to think of. And I think as of a few days ago, we were up to nearly 400,000 page views on that blog.
00:35:53
Speaker
on our site, which is an incredible reach. And I think that's where we can try to encourage with the basic articles. And it's not getting into a deep dive about all of the lovely conceptual stuff that we might debate amongst ourselves, but where we're offering practical advice to people. And I'm not going to name the news organizations, but I have had some success in encouraging a major print publication and a major broadcaster to change their
00:36:23
Speaker
their maps and I can't tell you how gratifying it is when you wake up and your social media goes wild and I'm thinking, right, has the world ended yet or can I survive another day? And really all you're looking at is, you know what, these guys made their maps better and that means that
00:36:41
Speaker
X hundreds of thousands of people are going to see a slightly better diet of that information today than they did yesterday. And it's these little steps, these little gains I think is where we can make a huge difference inside our community just to share that basic knowledge. And then maybe six months down the road when this is a fantastic teaching moment and a pivot in, you know, data sets that are being shared rapidly and how we can, how we're going to be able to look back on
Pandemic's Impact on Trust in Science and Experts
00:37:09
Speaker
it. Well, I hope we are.
00:37:11
Speaker
and use its teaching moment, then we can start to think about some of the detail about, well, is this graph better than that? And what about lines and curves and symbols and colors and all sorts of other things? So I have a lot of detailed things I wanted to talk about, but I think this conversation is moving towards the broader. So I want to continue that because I think we could probably all agree that over the last few years, there's been an attack on research and science and
00:37:40
Speaker
high-quality data, and I wonder, given the tone of this discussion and the tone of what's been coming out of, at least in the US, the federal government, and how even that tone has shifted over the last, even 48 hours, I'd say, I wonder whether you think this time is the return of the expert.
00:38:01
Speaker
And whether this is a pivot moment to turn maybe back towards more faith in science and an expert analysis, or you think this is just, you know, it's just a passing fad that we're going to get back to attacking science, attacking the media and attacking those, then generally we don't agree with. I will be happy to begin with that. So first of all, I would encourage our listeners to read Michael Lewis's The Fifth Risk, which is a terrifying
00:38:30
Speaker
book about fundamental institutions of truth telling are being undermined by the current US administration and what the consequences are of foolishly attacking experts and foolishly attacking people who know much more than we do about all this stuff.
00:38:48
Speaker
It's like, we live in society. Society is built on top of all these institutions. We better rely on them. All this populace bullshit about that we know better than anybody else, and we are all individuals, and whatever. It's like, no, we need to rely on those institutions, whether we like it or not. Even if you don't like it, you will need to. So that will be the first time. Now, I am not optimistic. I don't think that. I mean, I think that momentarily, at the moment,
00:39:16
Speaker
because there is so much fear and so much uncertainty, there's going to be a temporary return to let's trust epidemiologists, let's trust the government, let's seek help, right, let's seek help from the government. But as soon as the epidemic is over, right, certain sections, certain portions of the society, of society will go back to their own
00:39:38
Speaker
You know, they're all saying, so, you know, again, not trusting experts, you know, assuming that, you know, expertise means nothing and so on and so forth. And I hope that I will be proven wrong, but I'm not optimistic. Although I work towards avoiding that, but I'm not really saying that these societal patterns will change anytime soon. Amanda, what about you? Are you are you more optimistic?
00:40:01
Speaker
I want to be optimistic, John, but what was your follow-up question? No, I was going to follow up because you and I had a very brief Twitter discussion, I think, this morning about the need, and you've already talked about this, but the need for the information visualization people and the specialists, in this case, the scientists and those folks to come together more, right? There was this preprint that came out from the school, the school.
00:40:26
Speaker
the school that came out. The Imperial College. Right. And someone sent it to me and I was looking through it and I said, you know, wouldn't it be great if this red-green color palette wasn't red-green color palette? And that was our little conversation. So maybe I tweaked this question for you is not so much whether we're returning to the expert, but whether this is a pivot point where the date of this field does move where people start to rely more on the domain experts.
00:40:53
Speaker
I would love to see it happen. I know that, I also volunteer as the Operations Director for the Data Visualization Society. And one area that we've been looking at is, do we have an opportunity to help to connect data visualization experts with domain experts or small organizations or other researchers who are looking for a review or feedback or something else so that there's a constructive opportunity to provide data visualization expertise without
00:41:23
Speaker
putting things down in the world that aren't vetted by an expert. So that's something we're exploring right now. And if there's a way we can help to do that in our community, I know there's been a big call to action from our community of, I want to do something. How do we create a constructive outlet for all of that energy in a way that also makes sure that there's a little bit of validation of the skill sets that we're matching these organizations with? Because we want to add value, not to just create additional noise for them. I would say, though, I would add that on the optimism, John,
00:41:52
Speaker
I want to be optimistic, but I also know that I live in a bit of a bubble in the sense of being a community of database practitioners and a community of people who, for the most part, all have college, if not secondary, degrees. And when I look at the information that's landing with people and I look at who's listening to things, I mean, I reshared Newt Gingrich's Fox News post about the fact that coronavirus is a real problem because
00:42:20
Speaker
Hearing it from the other side and from a news source that someone trusts is going to be the thing that gets through to people. You can't just tell them not to believe a news source that's been their go-to for a long time. I mean, my 80-something-year-old grandmother climbed on a plane to Florida last Wednesday because she heard from all of the sources she saw that everything was fine. So I think we have to think about where people are consuming information, and I don't think
00:42:48
Speaker
in different news outlets that everyone is saying, let's trust the experts. So while I want to be optimistic, I think we also need to think about kind of where people get their information and how we have an influence on amplifying expertise across the aisle and across different platforms. I would say, by the way, that coming to Florida is always a good idea. So I don't blame my mother. She had a great time. She flew back on Sunday, Alberta. She had a delightful time with my hands down there, elevating about.
00:43:19
Speaker
We can all quarantine at Alberto's house. You are all welcome to come over. Thankfully, she was not flying through O'Hare airport during all the craziness. She was flying out of the small airport in Ronford, Illinois, which is a much better and safer place to go in and out of where there's, I think, two flights a day.
00:43:36
Speaker
I just want to follow up with what Amanda said. You know, Amanda, you were talking about where people get their information. So, you know, if you have two different data visualizations showing slightly different things and ones from one source and ones from another, we do know from health communication research and other communication experts that people are more likely to, you know, follow the information in their trusted source, right, and to share that information. And, you know, for a long time in public health, we talked about this kind of all
00:44:06
Speaker
knowing credibility. This source is the most credible and everyone should listen to it. And over time, I've started to say, well, that's not always true. And in my opinion, credibility is in the eye of the beholder. So in other words, who is it that you're trusting? And so I don't know what this means for what comes next, but what I am trying to do myself, and I have seen people doing this also in my social media feeds, is at least trying to share information that
00:44:34
Speaker
is neutral. In other words, trying to get information out there that, okay, here's this graph and I'm explaining why I think it's a good source to look at. And so trying to get people out of their information bubbles a little bit. And I think people are a little more willing to do that right now because they are anxious and there is this sense of panic about what's going to happen. And so I think people are really eager for information. And so I think it's a good way to say, okay,
00:45:02
Speaker
Let's compare, you know, here's different information pieces and let's see, okay, sometimes we don't know the exact answer. Sometimes there's different versions of an answer, right? And how we construct this graph or analyze this data can change that. And so at least I do feel like people are getting a little more comfortable looking at a wider range of information than they normally would. And so
00:45:24
Speaker
Whether that has any long-term effect on listening to the science more the experts more I'm not sure but I'm at least pleased that I see a little bit more diversity in terms of the sources people are looking at and sharing. Let's not also forget them yesterday a lot of people got their information from Arnold Schwarzenegger in a terminated t-shirt feeding carrots to a small horse and a mule.
00:45:47
Speaker
And that's where people are getting their information from. I think half a million people have either retweeted or shared it or liked it or whatever.
00:45:55
Speaker
You know, if Arnie's getting the message across, great.
Coping Strategies During the Pandemic
00:45:59
Speaker
That's the thing, right? It's like as long as the information is truthful. And I really like what you're saying about who the conduits of the information are. I don't care whether it's Newt Gingrich or what, and I hate Gingrich. I didn't know he has been a source to this country and a really bad influence. But in this particular, historically speaking, but if right now Gingrich is doing the right thing,
00:46:20
Speaker
I would encourage, I mean, I don't care just because I know that a certain portion of the population will trust that guy as a conduit of information. So whenever he says the right thing, we better promote that because it will reach certain bubbles that we don't reach to normally. Same things like if, I don't know, if Rush Limbaugh, for example, started saying, you know, things that make sense, we will need to promote that, right? I don't know that it sounds a little bit perhaps
00:46:48
Speaker
I don't know, disingenuous or something, but I think that we need to be pragmatic with all this stuff, right? So we've covered a lot, I think, and I want to end with not too quick, but just a quick personal thing. What are you doing to manage your time, your household, your lifestyle? We've all got different
00:47:10
Speaker
people in our houses at different ages with different needs and so if you're comfortable sharing something that you're doing that maybe maybe won't help other people but just to feel like we're a community again because it does feel like there's no one else around.
00:47:29
Speaker
So maybe we just go around real quick and something that you're doing or done or something that is helping you get through this, I would say, strange time. I don't know, Alberto, you wanna start?
00:47:42
Speaker
Yeah, sure. So I guess this is true of everyone who has small children that we are all homeschoolers right now. But finally, I'm sort of enjoying the experience, particularly with my little daughter, like sitting with her and going through the lessons in the morning. And, you know, the other day was there was someone tweeting that that school teacher should make a million dollars a week.
00:48:05
Speaker
And I completely agree with that, but at the same time, the experience of doing it has been refreshing in some sense. It's done consuming, but it's worth it. I think it has been a lot of fun. But then try to reserve some time for yourself also, away from social media, away from the news, and just do something that sort of like recharges your batteries.
00:48:28
Speaker
just meet with friends if you can, small groups or sit with family, do some reading. If you enjoy solitude and reading, just do that. Try to meditate. Meditation really helps. Mindful meditation, if you know how to do that, I would really encourage you to do that just because we are all going through a lot of stress. And I'm saying this from the point of view of someone who is privileged enough
00:48:54
Speaker
to be a tenured college professor. So, you know, I am in a different situation than someone who, for example, is fearful of losing their jobs, for instance, at this point, right? And I fully sympathize with that, with that sort of anxiety. I am not going through that. But I think that all these techniques can help anybody, try to cope a little bit with what's going on and what is to come, which I fear that is going to be a little bit, even a little bit worse. Jen, what about you, things you're working on?
00:49:23
Speaker
So I have two energetic boys at my house. We don't have a lot of space in our home and both my husband and I are working from home. So it's been interesting. Yesterday they were lucky enough to get a vacation day from school.
00:49:39
Speaker
able to do whatever they wanted because I had to start getting my classes organized and reaching out to my college students who are now having their lives upended and, you know, not able to come back to campus and having to move everything online. So I wanted to make sure I was getting to them and helping ease some of their anxiety about what's happening with the transitions.
00:49:59
Speaker
So my kids were very excited to have a vacation day. I do everything in spreadsheets. So we've already got multiple spreadsheets going with various things around our schedules. But at the same time, you know,
00:50:12
Speaker
I do that mostly because one of my kids really likes structure and needs the structure. But I'm actually really looking at this as an opportunity for my kids to explore things that they don't normally get to do. So actually before I left to come over here, I said to them, okay, I want you while I'm gone to think about something you want to do an independent study on and next Friday you're going to do presentations back to me. And they were so excited thinking about their topics. They were already starting to do some research.
00:50:39
Speaker
You know, I'm going to teach them some things that I haven't had time to do like gardening or things like that and make it into a learning opportunity and also sharing some of my knowledge about public health with them. We're going to be doing some different public health related projects to help them learn about what's going on and also at the same time learn about what public health is and so
00:51:01
Speaker
Again, I speak from a place of privilege where I'm able to work from home and have some flexibility with my schedule. So I really empathize with people right now who are in a tight spot. So I'm thinking of everyone and I think it's tough for everyone in different ways. Yeah. Ken, what about you? I'm going to start reading a pile of books that I've got somewhere.
00:51:29
Speaker
people. No, I'm actually, I'm actually finishing off writing a second book. So I've been I've had my head down kind of in my own little bubble for a few months. Anyway, I just I'm doing that at home rather than I guess at work. But the irony today was today was the first day that our company kind of mandated everybody could work at home if they really wanted to, which placed enormous pressure on our VPN. So everyone was working at home and there was no
00:51:54
Speaker
really good, strong signal to work through a VPN. So I've actually had to come into the office to do this. And it's lovely and quiet here. I might, I might stay here. But my, my partner also works the same company. So we're both working at home. She used to work at Imperial College in London, in the epidemiology department. So we're having some interesting conversations. The dog wants to know what's happening. Why are we around?
00:52:21
Speaker
But, you know, I think Alberto said making time for yourself, do different things. I mean, you know, we're baking fresh bread, you know, just make it yourself. Rationing toilet roll, obviously. That's got to come to an end, hasn't it? What is the panic buying with toilet roll?
00:52:36
Speaker
That's the whole of the podcast. You know what, if you want to do a fun visualization about the coronavirus that may not have, you know, implications or bad consequences, I would actually visualize the say of the toilet.
Community Support and Resilience
00:52:56
Speaker
Google searches for toy, when I buy toilet paper. It's the potential data, Alberto. Everyone needs to channel their energy into looking at all that peripheral data because that's where there are fun stories to be told about life or death consequences. Yep. Amanda, you kicked us off. So I'll give you the last word. You have a real little one at your house. So what's keeping you going right now?
00:53:19
Speaker
I know homeschooling, primarily playing with our current lexicon of favorite toys and enjoying the fact that a foolproof way to make him giggle up worsely is to throw him enthusiastically into a pile of pillows on the couch. So that's a fun game and I think good for some tricep work at the same time. You're not going to the gym. I also cook a lot and so I'm taking some kind of joy and reprieve and the fact that I can actually make good use of a lot of the stuff that's been in my pantry for a while.
00:53:48
Speaker
and spend some time going ahead and taking space from electronic media in different ways by getting in my kitchen and baking or cooking. So I'm not letting the volume of dish coming from that. Great. Well, thanks to all four of you for getting together for a bit. I really appreciate it. This was a really interesting topic, really interesting conversation. So take care. Thanks a lot. Stay healthy. Stay safe. Thanks, John. Thank you, John. Thanks, John.
00:54:19
Speaker
And thanks everyone for tuning into that episode of the podcast. There will probably be more coming your way as this pandemic unfolds and the data continues to change and update. I'm also very excited to bring you some of the video live chats over the next few weeks. So please do check out the website to get more information about those, when they'll be occurring and the Zoom links that you'll need.
00:54:45
Speaker
and feel free to bring your kids onto those episodes if you are home with your kids I'm hoping that they will be instructive and they will be informational and we're doing them on video so we'll be able to actually show some things so we're not restricted to the audio format that we have in the podcast so again I hope you're well and safe so until next time this has been the policy of his podcast thanks so much for listening