What COVID can teach us about the power of data visualization

[00:00:12.700] Liz Fraley

Good morning everyone, and welcome to Room 42. I’m Liz Fraley from Single-Sourcing Solutions. I’m your moderator. This is Janice Summers our interviewer, and welcome too Sarah Doan, today’s guest in Room 42. Dr. Sarah Doan is an Assistant Professor of Technical Communication at Kennesaw State University, where she teaches data visualization, information design and health and medicine in technical communication. Her research on instructor feedback has appeared in IEEE transactions of technical communication and her research on Covid-19 charts is appearing this month right now in the Journal of Business and Technical Communication. Today, she’s here to help us start answering the question, What can Covid Teach us About the Power of Data Visualization? Welcome.

[00:01:01.420] Sara Doan

Hi, it’s great to be here. Thank you so much for having me.

[00:01:04.780] Janice Summers

We are so pleased that you’re here with us today, Sarah. And what a timely topic when we’re talking about graphics and representation of information and what is it that we’ve learned over this last year? In a nutshell, well, let’s start unpacking it, because I know there’s a lot of information, a lot of use and abuse, right?

[00:01:34.510] Sara Doan

Yes, there is a lot to unpack here about charts and graphs. So one of the big things that has been ever-present during the Covid-19 crisis, especially in the United States, is the interplay between data visualization, as far as data literacy goes and as far as visual design goes, those two things need to work together to create quality charts. And we’ve seen um, I know it’s clich at this point to say a pandemic of information, I think Ken Burns just came out with something about that yesterday or today. But there really has been a pandemic of information where people are, you know, maybe creating these charts, maybe sharing these charts without really taking a look at what the data is either implying or what the visuals are implying. So there need to be those two literacies, the literacy about the data and the literacy of visual design that work together to create quality ways to share information about Covid-19.

[00:02:40.510] Janice Summers

Right. So they’re just taking what, the wrong conclusion from the data because they’re not interpreting the data correctly or they’re trying to make things seem a little bit worse or a little bit better in the visualization, what are some of the disconnects that you’re seeing?

[00:03:00.650] Sara Doan

So I will preference this by saying that in all of my looking at charts on the Internet, which I spend quite a bit of time doing these days.

[00:03:08.790] Janice Summers

Right.

[00:03:09.710] Sara Doan

I have never seen a chart that makes Covid look worse than it is. There might be examples out there, but over the past, however many eternities we’ve been spending with Covid-19 now, I have never seen a chart that makes Covid look worse than it is.

[00:03:29.090] Janice Summers

Right

[00:03:30.680] Sara Doan

I have seen a lot of charts, which some of them I wrote about in my publication for the Journal of Business and Technical Communication that minimize Covid-19 either through the data or through the design elements that are used.

[00:03:46.440] Janice Summers

Right.

[00:03:48.460] Sara Doan

So when we’re making comparisons across different charts that are meant to be viewed as one holistic unit, for example, there was a chart that grouped Covid-19 visually with the seasonal flu, as if to say Covid’s just like the flu, it’ll be fine. And this was in, I believe March 2020. So right when things were starting to get bad in the United States and we were starting to go into lockdown. And it just hasn’t been the case that Covid-19 is the same amount of severity as the seasonal flu, which is also a medical condition that people tend to underestimate in the United States. Oh, it’s just a flu.

[00:04:31.090] Janice Summers

Isn’t that also an example of apples and oranges and trying to make them equal the same because an apple is an apple and an orange is an orange, they’re not the same thing.

[00:04:41.950] Sara Doan

Correct.

[00:04:42.460] Janice Summers

So do you find that also happens in chart misrepresentation? Right or wrong you’re trying to compare two things and say they are similar, but they’re very different.

[00:04:54.130] Sara Doan

Yes, so in those pie charts that I mentioned that were created by Time Magazine, that’s exactly what happened. They’re trying to compare apples and oranges in data, and it just doesn’t work like that. And when you make those kinds of comparisons, it can be really misleading because people read the chart expecting to find, you know, a true representation of reality, when really we need to remember that charts are always interpretive. You can make the data tell or do any message as long as you manipulate it, which is not to say that charts can’t do a great deal of good in illustrating an argument. But I think additional caution is needed, especially now that programs that are wonderful, like Tableau or like even Excel, Microsoft Excel can create these charts very easily and then they can be spread very easily via social media or via news media that might not always be trustworthy. So you know, again, it’s another clich, but these charts have really started to go viral because people can see them on social media. People like to simplify arguments and understand what’s going on. And charts are one way of shorthand for doing that, and they can be used in these really amazing ways, like for example

[00:06:19.540] Janice Summers

Sorry, but we as consumers of content, consumers of information, because we can take on both roles here, we can create, but we are also consumers. We’ve got a lot going on in our lives.

[00:06:30.490] Sara Doan

We do.

[00:06:31.600] Janice Summers

So, a lot of times I rely on quick, easy things that I can capture like visually and hear so that it sinks in better for me. So it is really easy, that’s what makes them so appealing, right?

[00:06:48.160] Sara Doan

Oh, absolutely. And during the pandemic, there were two charts shared in March and April of 2020 that really changed my mind and made me realize, oh, this is going to be big and I need to prepare for lockdown yesterday, even though I knew it was coming back in late February because I was starting to read what was coming out of Wuhan. So there was this wonderful set of charts done for Reuters International about patient 31 in South Korea. South Korea’s contact tracing was so good that they managed to trace the first 30 cases of Covid in the country and then patient 31, an older woman thought she was positive for Covid but refused to go get tested and did things like going to her church for services that lasted multiple hours and then went with a friend to a hotel buffet and wound up starting a Covid cluster in South Korea. That was over five thousand people who were infected because of her actions, and so these wonderful folks at Reuters did this beautiful visualization where they go through and they trace her actions on a timeline, they’re like, this is when she was probably sick, this is when she should have gotten tested, this is when she finally did get tested and stopped circulating in society.

[00:08:15.590] Sara Doan

And then they’re like, here are all the clusters that happened because of her actions. So to see that in charts where they were depicting all the people who were infected with these little people icons was really powerful. Because five thousand of those fills up your whole screen. In a second example, Stephanie Jolly, a food scientist in New York City who is from, I believe, Kentucky or Tennessee, charted the effects of a Covid-19 mitigation in Kentucky and in Tennessee. So she used the state’s school basketball colors of this royal blue and then this orange, and she went through and charted how many people were getting tested versus how many people were getting infected in mid-March 2020. And then she was able to tie infection rates through, adding annotations to connect infection levels to how governors were reacting, inputting these protections in place for people by doing things like canceling school, by doing things like saying, you know, the right to worship is part of the First Amendment, but you shouldn’t be killing people by doing it.

[00:09:34.970] Janice Summers

Right.

[00:09:37.260] Sara Doan

So she was able to tie all of the infection rates that differ drastically between Tennessee and Kentucky. Because Tennessee was much less cautious than Kentucky was, so they had higher cases and those charts went massively viral on Twitter, she ran occasional updates for about a month. But it’s just been so exhausting that unless it’s your literal job, it can be hard to keep up with.

[00:10:03.920] Janice Summers

Yeah, it becomes very tasking. And you mentioned she used the state colors? Okay.

[00:10:12.510] Sara Doan

I believe it was royal blue for Tennessee and then orange for Kentucky, because those are the big flagship university colors.

[00:10:22.170] Janice Summers

So she used the colors to help the people identify this is my state.

[00:10:27.460] Sara Doan

Exactly. You know, there was some use of design tools to help people kind of take ownership of what their state was doing because we royal blue folks identify with the royal blue team. I’m sorry, I haven’t lived in the South long enough to internalize which team is which.

[00:10:45.510] Janice Summers

You know, I never went to University of Michigan, but I know the blue and gold, right.

[00:10:54.520] Sara Doan

I can do the Midwest, that’s about it.

[00:10:56.260] Janice Summers

I was born in Missouri, I live in California, I’ve lived here for years, but I recognize the blue and gold, and I also recognize the Spartan green, hats off to Michigan State.

[00:11:09.160] Sara Doan

Take away one from this talk, people love color coding things.

[00:11:12.910] Janice Summers

Yes, it’s another one of those visual cues that we use, right.

[00:11:21.730] Sara Doan

Precisely.

[00:11:23.660] Janice Summers

So what are some of the things that we’ve learned that help us to develop better visual, like better visualization, saying we just talked about color code and its always accessible for people, so we have to take that into consideration.

[00:11:40.040] Sara Doan

Right. So testing things in greyscale is very important. If the visualization doesn’t work in grayscale, then it’s not going to work out in the world.

[00:11:50.660] Janice Summers

Right.

[00:11:51.440] Sara Doan

That’s a pretty good rule of thumb. Another thing that I’ve been paying close attention to is the connection between visuals and the verbals on a visual. So when I mentioned the chart about Tennessee and Kentucky’s Covid policies, that chart would not have been as effective without the verbal cues, what Alberto Cairo calls the annotation layer, a strong title that gives the kind of high level wisdom of a chart and then any data labels. 539 Covid cases on March 10th. Those all need to be labeled because people remember visuals much better, or actually I shouldn’t claim that because I don’t actuallyI cannot scientifically prove that, so I’m not going to claim it.

[00:12:43.310]

But one of the things that helps people understand visuals is actually the words. So even though it sounds kind of counter-intuitive, a strong annotation layer is what keeps things from getting lost, and especially when things are circulating on social media, It’s so important to keep all the words with the chart. And Governor Cuomo who did all of those wonderful media briefings in the state of New York, actually had a problem with one of these charts that I also write about in the JBTC article where his office created this beautiful chart talking about net hospital admissions in New York City, and his speaking over the chart during the briefing was very, very clear.

[00:13:30.080] Sara Doan

But those same words weren’t necessarily present on the chart when it was circulated by news media afterwards. So it looked like there could have been only 200 more admissions to hospitals in New York City, there also wasn’t a good sense of what counted as a hospital because that was back during the big surge in New York City. So are we counting emergency medical centers, does this include people who might not have the means to access treatment, were they turning people away from the E.R.? Saying that hospital admissions only two hundred more. That doesn’t really tell the whole story, so when we’re making charts and making these arguments, we just have to be very careful that if things are pulled out of context, some of the context needs to come with or be readily accessible. Another thing that we’ve learned throughout the Covid-19 pandemic is cite your sources. Just saying this is from the CDC isn’t good enough anymore, because that website, as much as I love it and as much as I fangirl over the CDC here in Atlanta, the website has some navigational constraints.

[00:14:50.900] Janice Summers

Yeah.

[00:14:51.890] Sara Doan

So it’s not always easy to go back and find the same information. So being very conscious of where information is coming from, who is producing that data, you know, who is representing that data becomes very important as well.

[00:15:05.390] Janice Summers

Well, and I like people that cite sources and it’s fine you know, I watched Cuomo a lot, especially in the very beginning and he always had this just the facts, these are the facts. This is Deliver all this information factual right, citing your sources adds credibility, and if you’re not citing your sources, then you lose trust. I think then you’re just into a world of sensationalism. So you need to be able to back it up. Now, when you talk about data of words going along with the pictures, words and pictures in harmony so that they’re not separated, how would you do that? Is it like you’d have like, well, I’m thinking of like technical illustrations and how they have you know the bill of materials has the parts list and everything points to something. Is it kind of like having that type of dynamic with a graphic? How would you do that? How would you attach data?

[00:16:08.330] Sara Doan

That’s a great question. I would say not quite, because if I were doing a one to one of a technical illustration to a data visualization, that would mean labeling all the parts and saying this is a bar chart or this bar shows how much stuff there is. So it’s not quite a one to one relationship of technical illustration to data visualization. In this, we’re labeling the data, not necessarily the visuals.

[00:16:35.330] Janice Summers

Right.

[00:16:35.330] Sara Doan

The visuals are there to show, the words are there to explain and to prevent misunderstanding to the extent that it’s possible. So instead of, you know, look for an example on my little cheat sheet here, you know, if you are charting a line graph that goes up and down of Covid-19 cases, you’re going to want to put a data label on every single instance of that every time you count cases.

[00:17:13.580] Janice Summers

Right, data point.

[00:17:13.580] Sara Doan

You might have 33 one day, you might have 61 the next day, you might have 86 the day after that.

[00:17:18.950] Janice Summers

Right.

[00:17:19.430] Sara Doan

And so you want to make sure that you’re being really transparent how that’s being labeled.

[00:17:30.240] Janice Summers

Okay.

[00:17:30.240] Sara Doan

And that transparency and those verbal elements, so the word elements also contribute to it’s better accessibility. One of the things I’ve been interested in and I did a little bit of research with some great folks at Iowa State when I was there for my Masters Degree is Captions. Technical communicators don’t always write captions as robustly as they should. And those are really important for blind users, for users who are using screen readers. And just like I said before, to prevent misunderstanding. So that alternative texts, that captions or those captions, anything we can do to increase our audience here and to enact good universal design principles is time well spent.

[00:18:18.900] Janice Summers

Right, and one of the important things that, you know, it’s easy to forget. I mean, I think I was conditioned years and years ago, never forget your Alt text because it’s important, right, it has a lot of meaning even for people who You know, because if you have a lot of things that are very visual online, there’s people who just don’t like a lot of visuals and they just want the Alt text. They don’t want to see a picture, they just want to read what that picture is supposed to be about.

[00:18:46.680] Sara Doan

Right.

[00:18:46.680] Janice Summers

It’s easier for them to chew through information that way. So I’ve always been trained Alt text, Captions are key.

[00:18:58.830] Sara Doan

Right, and I’ve also been thinking a little bit fairly informally about helping neurodivergent users. How can we get people the information we need without overwhelming them, and then Alt text also becomes an issue of literal bandwidth. If my computer can’t load a visual, but it can load the Alt text, that is really important especially in this weird age where people might be relying on their Internet much more than usual to get necessary information.

[00:19:35.310] Janice Summers

Right, so a good alternative helps them understand, even though you don’t necessarily have to have it, it could be just that I want you to understand the gist of this image. If it’s like an emotional thing, I just want you to understand the gist of what this image is. Now, how about like on a global scale? We were talking about minimalism, clean writing, let’s talk about this on a global scale as well. How does this impact the audience that you’re speaking to? How do you make these graphs universal?

[00:20:14.390] Liz Fraley

Yeah.

[00:20:16.160] Sara Doan

I think, that’s a tough question in some ways, because so many of our visual design cues and our grammar for creating data visualizations is very centered in Western or let me rephrase. The communities I have been involved in have done a lot of data visualization in Europe and in the United States and Canada and Mexico, so that’s what I can speak to. I don’t ever want to erase the Knowledges or erase the data visualization tools used by people in other parts of the world.

[00:21:04.110] Janice Summers

Right.

[00:21:06.180] Sara Doan

Which is not to say that this way is absolutely better, because I’m sure that there are really interesting and useful and good ways to visualize data that I’m just not exposed to.

[00:21:16.620] Janice Summers

Right, well and then so there’s this thing of, you know, because sometimes you need to localize your graph if you want to reach a specific audience. You need to cater to how that audience communicates. So those things might be applicable to one audience and not another, is that true? I mean, when you’re talking about medical information, it’s really important that you’re targeting who you’re targeting, right?

[00:21:42.810] Sara Doan

Correct.

[00:21:44.860] Janice Summers

Because the point Isn’t the point for them to understand, so that becomes the critical part, right?

[00:21:51.720] Sara Doan

Right. And I do think that good design is somewhat universal, but I also don’t want to speak for cultures that are not my own.

[00:21:59.760] Janice Summers

Right.

[00:22:00.660] Sara Doan

And I do know that Japan has a long tradition of good information design. And even me using the word good is somewhat loaded here, rather, it’s information design that I respond to because it relies a lot on visual cues that I pick up because of the training that I have, but, you know, I just don’t have the expertise to speak on cultures that are not my own in this instance.

[00:22:32.090] Janice Summers

Right.

[00:22:34.590] Sara Doan

You know, all I can do is do my best to do what I think is universal design, and if there are better ways, I definitely want to learn them and respond to them and teach them.

[00:22:46.670] Liz Fraley

Yeah, so I was going to come back because I’m still back on the couch. So it’s interesting because sometimes it’s captions, a lot of the times professional and technical communicators call them titles. The title on the figure rather than the caption on the figure. And you’re right, we don’t always write them well, right, like, here’s the table about whatever. It’s like yeah, yeah, I can see it’s a table and I am gonna be reading about this the whole time, so I’m pretty sure that’s part of it too. Any tips for making, like, really writing good captions?

[00:23:28.940] Sara Doan

That’s a great question. So I would say that Titles are different than Captions, those are two separate things. The Title is the big picture thing that’s happening, new Covid-19 cases per day, for example, whereas the Caption should be explaining what’s happening. Caption should be using verbs, sometimes titles can do that, and it does vary depending on how much space you have where this is being shared, it’s very contextually dependent, but I would say that in a data visualization that I tend to champion, there’s a difference between the title, which is the bare bones of what’s happening and the caption, which gives a little bit more depth. One place that does that really well is in the Reuters article that I referenced a couple of minutes ago that goes through and does 6 different charts about patient 31 and that giant outbreak in South Korea back in February 2020. They label each chart with a title and then they have explanatory text underneath. So I think including both is great, especially for neuro-typical users oh sorry, neuro-atypical users.

[00:24:54.590] Liz Fraley

It gives me an idea for a follow- up guest, because I know someone who works with neurodivergent, p eople who are looking for work and other things

[00:25:04.370] Sara Doan

And, sorry to interrupt but neurodivergent is the word I should have used there.

[00:25:10.260] Liz Fraley

Okay. I wrote it down that’s why I got it right. So while we’re on that topic, anything that we should, like that are easy, I know this is not the right way to say it, but like easy things that we can do immediately, that we just, that if we’re not used to working with neurodivergent users or thinking about them, as none of us like, unless we’re very careful, we have time constraints and we’re short right, what are the key things that we can do right away that we should be aware of?

[00:25:47.630] Sara Doan

I would say, information should still be complex, definitely I don’t want to talk down to people, but presentation should be streamlined and it should be well thought out. I never want to say simple, I don’t know that I typically fall into the minimalist camp myself as a designer, but making sure that what you want people to know is emphasized is really important. Keeping structures the same, paying close attention to how information is structured on the page, on the screen. This is not necessarily a data visualization example, but in my teaching, for every single class I do, I include a checklist where these are the tasks you need to do on which day and students respond really well to that because every week is structured the same. You know, right now, information is so overwhelming and Covid-19 is overwhelming and the political context that we’re in is overwhelming and making sure that information is usable and making sure that information is well-structured is really important for all users. It’s good universal design.

[00:27:07.610] Liz Fraley

So patterning to give it away, thinking about it when you’re making multiple visualizations or multiple charts, do you want to have sort of that same patterning in what you’re doing for whatever it is you’re delivering, not necessarily across everything you produce, but within whatever it is that you’re producing?

[00:27:30.940] Sara Doan

Yes. You should have some sort of visual grammar, you should have a stylesheet. Again, I’m going to talk about this Reuters chart till the end of time, all right, you know the Reuters article that I’m referencing, does it really, really well. You can tell that a unit of one pictograph of a human is one person, and it is usually blue unless it is red for a specific reason. So they use red to represent international travel in this set of infographics and data visualizations, whereas blue is for people who did not travel internationally. So there are definitely ways to use color and to use design to humanize the data, which is something we haven’t talked about yet.

[00:28:23.120] Janice Summers

Right.

[00:28:24.100] Sara Doan

One of the things that I mentioned in my JBTC article is that reducing Covid-19 down to a binary of life or death totally erases about the one-third of people who get Covid and become long haulers. We’re going to see cardiac issues, we’re going to see lung issues, we’re going to see pulmonary issues for a very, very long time because of Covid-19. And I think that reducing things to the binary of you survived Covid-19 versus you passed away from Covid-19 is a really dangerous elision of that in-between space. It is not so simple in reality. And that’s also something that a lot of disability rights activists have been talking about as well. And that also becomes an issue of race and class.

[00:29:16.690] Liz Fraley

Yes.

[00:29:17.200] Sara Doan

Because so many of the people who have died from Covid-19 have been African-American, they’ve been Native-American, and there are a lot of white communities who have been virtually untouched. And that becomes a big social justice issue, and there are people who are much more qualified to speak to that than I am, but one of the things that I’m noticing in data visualization is scrapping out that human element, and sometimes the prevailing narrative, especially in March 2020, was, oh, white people are going to be fine. It is not fine to do that. So being able to complexify the data, to humanize the data is so important. And I was talking with my students about that yesterday, we were looking at charts and graphs coming out of Mississippi as far as the ratio of people who have died versus people who have survived Covid or Covid-19 cases.

[00:30:17.780] Sara Doan

And, you know, the chart visually minimized this ratio of cases to this ratio of deaths. How many people have passed away, and students and I were like, you know, this is 5000 people and they are probably people from marginalized communities. So it’s really important to keep that in mind as we think about Covid, as we create public health interventions around Covid, and as we continue to move forward in thinking about public health, who are the most vulnerable, and how can we design to make sure that we’re not just designing for upper middle-class white people.

[00:31:01.790] Liz Fraley

And that’s what I thought of as soon as you were saying, you know, when you reduce the in-between space, right, that’s classically how it happens right there.

[00:31:12.560] Sara Doan

Right?

[00:31:13.580] Liz Fraley

Yeah. Well, that’s so many to think of, I can’t I’m stopped like that I’m thinking. it’s alright, go ahead.

[00:31:24.710] Sara Doan

You know, it is really tempting to reduce things to a black and white kind of thinking and especially with Covid, and then I’m not even touching the whole class dynamic of grocery workers, so many grocery workers have gotten Covid, so many delivery people have gotten Covid, it’s just a huge problem.

[00:31:50.130] Janice Summers

The Farming community I think, in California, Imperial Valley, which is a lot of farmers, a lot of pickers.

[00:31:56.800] Sara Doan

Right.

[00:31:57.820] Janice Summers

Right.

[00:32:07.330] Liz Fraley

Not to bring us out of the seriousness, although because I need to think on all of that

[00:32:13.570] Sara Doan

Right. I’m not sure how I can top all of that.

[00:32:15.690] Liz Fraley

Well, it’s almost bigger than I can conceive at the moment. And we’ve talked several times, again I’m still cut short. So let’s spin a little bit and talk about I know that you’ve talked about things that go viral and this and that and that there are things that can sort of push things that way, whether for good or for bad color or words or the way things are laid out, can you give us a little bit about that.

[00:32:53.900] Sara Doan

Without trying to talk too much about politics, this is the day of the inauguration, one of the examples that I look at in my JBTC article is from Fox News in Colorado, Fox 31, and they just butchered this chart and it went viral in the bad way on Twitter, someone tweeted, You can teach an entire semester of how to lie with statistics on the y axis of this chart because it tries a logarithmic scale and then starts randomly jumping numbers to minimize visually Covid-19 cases. The area between cases, 33 cases and let me see 129 cases on the Y axis is actually a greater visual distance than the difference between 174 and 344 because they butchered the scale that badly. So I think that even though people love to trash social media and much of that criticism is warranted, it’s still just so interesting how the news media is using charts in this way for very partisan ends that don’t benefit humanity unless they are part of this in group.

[00:34:27.130] Sara Doan

So I think it’s very telling that charts don’t just go viral by themselves. There are human actors, there are robot actors, not in the conspiracy theory sense, but in the Twitter bot sense, and it’s just been so interesting to watch misinformation explode.

[00:34:55.700] Liz Fraley

And the compelling part of that was, look how much it jumped and how much it isn’t jumping right they’re looking at the rise and fall. And rise and fall is one of the key indicators for us when we look at a chart.

[00:35:11.420] Sara Doan

Right, and you know the way that they structured this chart, you can visually look at and say, oh, it’s not that bad. Oh, no, it is very bad, it is very bad. And I have just kind of been screaming into the void about this since March of 2020 where these things are bad, we need to take them seriously. You know, I spend a lot of time thinking about public health interventions and how can we better communicate CDC guidelines to people, because, you know, if you’re exposed to Covid, it’s not always clear what you should do. So I don’t want to wander too far away from the question

[00:35:57.820] Janice Summers

Sorry, so then as designers of content that we know this content is going to be consumed, right, we need to I think, probably focus more on how to integrate hard data points into the image, so that doesn’t get Or when it does go viral, that the data accuracy stays intact with the graphic visualization, right?

[00:36:25.140] Sara Doan

Exactly.

[00:36:25.890] Janice Summers

Because sensational things are always going to be sensational and there’s always going to be bad players out there who try to manipulate things for their own end, and you’re not going to get around that. But as technical and professional communicators, that’s not what we do. And I think that we need to have that emphasis so that we can change, right, the way other people do, because you’re going to come to technical and professional communicators to generate those graphs and charts. So if we adopt some better design practices and some tips to make things, you know sneaking in those hard data points into it so that it doesn’t get into the hands of somebody who might be a bit mischievous and go share it off on social media. And then it’s all detached, right. You talk about Cuomo and he had a beautiful graphic, right, he had the visualization was accurate, but the data needed to go with it so that when it went viral, it wasn’t detached.

[00:37:30.630] Sara Doan

Right, and it was one of those things where when Cuomo was talking about it, it was perfectly clear.

[00:37:36.360] Janice Summers

Right, because he thinks of the data points.

[00:37:38.820] Sara Doan

Right.

[00:37:39.570] Janice Summers

Yeah. So making sure that stuff stays together, and that’s why I’m such a big proponent of the annotation layers, you know, that circles back to earlier in our talk where we were talking about how important it is to label all of the data and to put a good informative title and then to write a good caption if you’re able and that alternative text, because the more you can put with the chart, the less it can be redistributed. And for a long time, we could look at a chart and realize, oh, a professional made that. As recently as five years ago, it was really clear when a data designer made something versus when literally anyone made something.

[00:38:23.010] Janice Summers

Right.

[00:38:23.770] Sara Doan

And because I spend 60 hours a week looking at Microsoft products online or on my screen, I can tell when something’s been made in Excel. But not everybody has that same design knowledge. So, you know, one thing that’s been really prevalent and fascinating and frankly, depressing lately has been the number of people who are using high-end design strategies or things that they find packaged on Instagram or in Tableau, and we can’t necessarily trust how something looks anymore. You know, until pretty recently, you could look at a website and say, oh, someone who knows what they’re doing designed that I can probably trust it. But now misinformation is packaged to look just like regular information. The quality of design that’s being used is no longer a reliable indicator of whether something is going to be a reliable source. I know that they’ve had a lot of problems on Instagram with QAnon. Women sharing theselike brightly colored pastel visuals to get people into conspiracy theories. There’s been articles about it in the Atlantic, you know, some really smart writing has been done about it. But, you know, conspiracy theories look pretty now. It’s bad, because we can no longer use design to filter out misinformation.

[00:40:00.150] Janice Summers

Right. It’s easier for people to create graphics, I mean, people are even doing deep fakes in video and audio, so

[00:40:13.440] Liz Fraley

Well, even beyond that right, anybody can throw up a store or a website and look like they’re official, right, there’s no barrier entry to it hardly even at all

[00:40:25.440] Sara Doan

Right.

[00:40:26.430] Liz Fraley

And really finding out whether they’re real, an actual company, not just somebody overlaying something else.

[00:40:35.520] Sara Doan

Correct.

[00:40:36.700] Liz Fraley

Or selling the same product under a different name, even just because it sells better that way on this website versus another one.

[00:40:45.790] Sara Doan

Right.

[00:40:45.790] Liz Fraley

Like there’s a great story about Alex Jones and Gwyneth Paltrow I think, I can’t remember exactly, so I’ll stop there.

[00:40:54.040] Sara Doan

Yeah.

[00:40:55.040] Liz Fraley

Right.

[00:40:55.630] Janice Summers

Hurt some people.

[00:40:57.110] Sara Doan

I’m just going to sigh

[00:40:58.570] Liz Fraley

They were opposite, and they were selling the same product under a different name. The same product to their audience is kind of fascinating, but you can’t necessarily know that something is real. You can’t just by looking at it anymore, which is odd and the story is like distressing.

[00:41:19.760] Sara Doan

Yes.

[00:41:21.440] Liz Fraley

All right, so we’re coming on close to the end, but I want to go off a little to the side only because we’re talking You mentioned annotation layers

[00:41:29.570] Janice Summers

Yes.

[00:41:29.570] Liz Fraley

And graphics, so I was brought to the mind of the current thinking about presentations and slides, right. Which are visual depictions of your talk. There’s a bit There was a big push over the last, I want to say, 10 years, no words, just a picture which works great when you’re there to talk over it like Cuomo. It’s great when you’re talking over it. It’s not so great in reference, or the see later, but the graphics can still go wherever they’re going to go. So how would Annotation layers like are there ways to build that stuff in? Should we think about that stuff there, too?

[00:42:13.380] Sara Doan

Again, that’s going to be very contextual and very audience-driven, but if you’re putting up slides on the Internet, you probably want to include those notes in the notes section if you’re keeping things in Google slides or in PowerPoint. You know, there are ways of annotating you know, including comments on a PDF, for example, personally, I started hating on the pictures only slide trend before it was cool, because I need the verbal reminders when I’m talking even before the pandemic happened and took 90 percent of my brain away and replaced it with endless internal screaming. I like having the visuals and the verbals work together. And I think there’s probably going to be a big push to bring those back together, or at the very least, that is my taste and preference in design, because I did a lot of grad school and education theory has always been important to me, especially in my research on instructor feedback, and I found that the more you integrate the verbals and the visuals, both written words and spoken words, the better people pay attention and the better people remember

[00:43:35.090] Sara Doan

I need both when I am observing a talk, I want people to have both when they’re listening to me, I find that especially with students and their attention, I want to honor that by putting things on the page or screen in front of them while also speaking about it. And it’s just good accessibility practice anyway.

[00:43:57.250] Liz Fraley

As you were talking about it, that’s what comes to mind.

[00:44:00.250] Sara Doan

The design is Universal design. It’s all coming together.

[00:44:06.060] Liz Fraley

It is and you never know where it’s going to happen.

[00:44:08.940] Sara Doan

Right.

[00:44:14.130] Liz Fraley

Well Awesome, I think we are at time. This has been really great, I’ve got a lot of notes, so I’m going to send you an email so that I can put links to all of this on event page.

[00:44:25.970] Sara Doan

Okay, of course.

[00:44:26.250] Liz Fraley

And then anybody who needs, wants to go deeper. I want to go read all the charts I can now.

[00:44:33.300] Janice Summers

I’m going to start looking to see if people have annotation layers, because I think that I think that Sara, to me, that seems like that’s the best way that we can safety and future proof the content. Because it also gives you the chance to have the slice of time that it applies to because some data is time sensitive and you don’t want someone circulating a graph that’s from like 10 years ago. It doesn’t apply anymore, so we need to be time-sensitive. And in the day and age where people get a hold of something and it goes flying off and it’s unintended, right, but it goes flying off and all of a sudden they’re separating it from fact, then it becomes fiction, you know, that’s not good. And I think, to me, I think annotation layers in that discipline of doing and because that takes discipline to do. And I think that is another key discipline that we need to have when we’re doing charts and graphs and visualizations. It’s great to use colors and there’s a reason there’s a psychology for all of that. But as truth, because in technical and professional communications, we pride ourselves on truthfulness and the information that we provide, and that’s what we’re paid to do, and I think that annotation layer gives us that that level of comfort that we can be it’s a higher integrity.

[00:46:03.000] Sara Doan

I agree, and technical communication is all about ethics and accountability, and we’re also word people, you know, we should be uniting these design tools that we’re using with the words that we have to you know, this is going to sound very nave, but create a better future for humanity, which is why I do the work that I do. One thing I didn’t touch on, but I’ll talk about elsewhere is that in the second golden age of data design, weird charts are back because we have these technological affordances.

[00:46:37.970] Janice Summers

What’s back?

[00:46:39.290] Sara Doan

Weird charts.

[00:46:40.670] Janice Summers

Weird chart?

[00:46:42.020] Sara Doan

Weird charts, and so the annotation layer becomes important because you need to tell people how to read them like the one tornado chart of Covid deaths.

[00:46:50.480] Janice Summers

Yeah.

[00:46:51.740] Sara Doan

Oh, that was a time, April was wild.

[00:46:56.220] Janice Summers

Yeah.

[00:46:57.710] Sara Doan

So if you ever want to talk about weird charts, I am happy to talk about weird charts.

[00:47:02.040] Janice Summers

We’re going to have to have another Room 42 on weird charts, and I think on that one, we might have to like break the rules and do some screen share.

[00:47:10.820] Sara Doan

Yeah, I have so many examples.

[00:47:12.890] Janice Summers

We will have to book that for the future then. Weird chart day, we’ll just shred some weird charts. That would be a lot of fun, that would be so much fun, I could literally just do like a cool thing of weird charts.

[00:47:26.390] Janice Summers

All right, well, I’m going to take you up on that one because I think it would be a lot of fun too.

[00:47:30.920] Sara Doan

Perfect.

[00:47:31.760] Janice Summers

I always want to make Liz laugh. It’s always fun to make Liz laugh so.

[00:47:37.850] Sara Doan

Thank you so much for having me.

[00:47:39.680] Janice Summers

Oh, thank you. It’s been a pleasure talking with you. Some really key, important information for all of us. And I hope everyone got some really good tips out of this one, I know, I did.

[00:47:50.990] Liz Fraley

Yeah, I’m seeing all the audience, too. That was great.

[00:47:54.140] Janice Summers

Yeah.

[00:47:54.800] Sara Doan

Fantastic. Well, again, thank you so much.

[00:47:58.100] Janice Summers

Thank you. And enjoy the rest of your day, thank you for coming, thank you for being our guest, and I look forward to talking to you again.

[00:48:05.570] Sara Doan

Absolutely, any time.

[00:48:07.370] Liz Fraley

Absolutely. Bye, everyone.

[00:48:09.650] Sara Doan

All right, bye everyone, take care.