Become a Creator today!Start creating today - Share your story with the world!
Start for free
00:00:00
00:00:01
Episode #83: Chad Skelton
169 Plays8 years ago
Many people agreed with my basic argument during last month’s Month of Story. One person, however, had a slightly different take on how we tell stories with data and what that concept means as it is applied to data. Chad Skelton, formerly...
The post Episode #83: Chad Skelton appeared first on PolicyViz.
Recommended
Transcript
Introduction and Guest Background
00:00:11
Speaker
Welcome back to the Policy Vis podcast. I'm your host, John Schwabisch. Thanks for coming back and tuning in. On this week's show, I get to talk to someone who I've always wanted to meet, actually did get to meet, but when I met him, I had no voice. And so we didn't have a chance to actually record the podcast. So now we are doing it from far away, basically across North America.
00:00:33
Speaker
Still good enough because we get to have a conversation. I'm excited to have Chad Skelton on the show. Chad, how are you? Welcome to the show. Thanks very much for having me. Now, we have been communicating for a while, social media, emails, whatever, and then we got to meet a tapestry and that's when I had no voice. My recording equipment sat in my hotel room, but now we get to have a conversation.
00:00:55
Speaker
I'm excited. I think we're going to talk about two main things. We're going to start by being nice to each other, and then we're going to take off the gloves, and we're going to find it out. So we'll see. So for those of you who don't know, Chad is a Tableau Wiz instructor, currently an instructor at Kwantlen University in Vancouver.
00:01:13
Speaker
and the University of Florida teaching an online course and a background as a journalist.
Teaching Journalism and Data Visualization
00:01:18
Speaker
But Chad, I'm going to let you talk a little bit about yourself before we sort of dive into content. So why don't you tell folks about yourself and where you come from.
00:01:26
Speaker
OK, well, thanks a lot for having me. So I was a data journalist for several years at the Vancouver Sun, which is the main newspaper in British Columbia, Canada. I took a buyout from the paper in September of 2015. And ever since, basically, I've been doing various different versions of teaching. So my main teaching gig is I teach journalism and data visualization at Quantlin Polytechnic University, which is just outside of Vancouver. I teach an online data storytelling course at the University of Florida as part of their new audience analytics program.
00:01:56
Speaker
And then I do a variety of public Tableau workshops, on-site Tableau workshops, and do a little bit of consulting. But I'd say the vast majority of what I'm doing is some variation of Tableau training, data visualization training.
00:02:12
Speaker
Florida job? For your Florida gig? Yeah, for the new Florida course. So I'm offering it for the first time this summer. And I had sort of relied on a lot of people on Twitter to sort of help me to suggest various readings and videos and stuff like that. So I decided to sort of post it just because I sort of found it kind of helpful to have a bunch a long list of
00:02:28
Speaker
fun videos to watch and cool things to read. So people can find that on my website, which is chadskelton.com. Right. And so I'll link to that on the show notes so people can chime in. And I'm sure people have other recommendations for stuff that you can use. Let's start by talking about teaching a little bit. So you're teaching both live in class and then online classes. So I guess the question is, how do you feel about teaching data vis in these two very different platforms?
00:02:57
Speaker
Right. So, yeah, I mean, the online stuff is pretty new for me. I mean, most of my training has been in person. It's been teaching sort of an undergrad class at Quantlin and then doing these sort of two day public workshops and two day onsite workshops. The University of Florida course this summer is going to be my first time teaching an online course. It's been kind of fun to sort of think about how to put that together and
00:03:15
Speaker
And do things like that and I'm hoping I haven't actually officially announced this on my website yet But this is a sort of exclusive to the policy biz podcast that I'm hoping in the in the fall to start offering my two-day Tableau Workshop in some sort of online format probably kind of three half days And it's just been kind of interesting because I find it both sort of exciting cuz I think with an online audience You can obviously reach a lot more people. I mean for now most of my students have been in Vancouver starting with this Florida course They're gonna be kind of all over the states
00:03:42
Speaker
But I also think it's a bit of a challenge because I when I've been sort of thinking about how I teach so much of it is just kind of scanning the room and looking for kind of those furrowed brows or those looks of boredom and you kind of know people keeping up are people kind of falling behind. And I think and I talked to a few people that have taught online about this. Like I think it's just going to require a little bit more deliberate kind of checking in with people like, you know, just type into your chat window. Are you doing OK? You're following along just to make sure that people aren't falling behind.
Engaging Students Online
00:04:08
Speaker
I think it could be a little bit harder to notice that. Yeah. Online would in person.
00:04:11
Speaker
Yeah, I think that the challenge, whenever I take a webinar, it's so easy to just slide the webinar video over to the side and check your email and browse the internet. So that's the thing. I think those check-ins are really useful. Some of the tools have breakout rooms where you can give people a little homework assignment or something or whatever. But again, one of the things that's great about data visualization or learning, or in our case teaching data visualization, is
00:04:38
Speaker
engaging in these conversations and these debates about, well, I like this chart, but you don't and why and how would we address some of the things that we do or don't like? When you're teaching in class, how do you get people engaged so that they're excited about having these arguments and debates? Right. I mean, what's interesting about the course I teach at Qualen being an undergrad course in the journalism program is
00:05:02
Speaker
students are essentially forced to take it. And journalists are sort of notoriously for good reason bad at math and they're not super excited. And I actually started asking students in my end of your kind of survey, you know, at the beginning of this class, how excited were you about, you know, data visualization and the bars are very much on the
00:05:23
Speaker
you know, one out of 10, two out of 10, three out of 10. Luckily, I moved them to the right, but it's actually kind of a fun challenge because it means you're kind of trying to sell people on something that they're not already convinced by. It's very different than my Tableau classes where people are paying a significant amount of money, they know they're already really into data visualization, their boss is told and they have to learn it.
00:05:41
Speaker
But I find both kind of exciting because it's sort of fun to sort of try to convince journalism students that love to write, hate math, that this is something that's actually exciting and they get interested in. And I think it's just one of those things where like a lot of teaching, you have to just remember how green your students are. And that you can sort of ask them a question that you think is obvious, like what do you think of pie charts? Like in our community, who doesn't have an opinion about pie charts?
00:06:04
Speaker
These students do not, right? So they need to have it sort of slowly introduced to them and kind of really give them a bit of time to think about it first before you start asking for their opinion on it because they're not going to, you know, off the top of their head have really strongly held opinions about data visualization. Yeah. Yeah. That's a great point. So Chad, in a lot of your classes, you're teaching Tableau. How do you marry teaching a specific tool like Tableau with core principles, you know, best practices that people need to get started in the field of data visualization?
00:06:34
Speaker
Yeah, I mean, one of the things that I found kind of interesting when I look at sort of the training that's out there is it seems to be broken down into sort of two main groups, there's kind of the concepts training, a lot of really talented people like Andy Kirk, and Colness Bomernaflek, and stuff like that, that sort of teach people the concepts of data visualization and how to think about data visualization without sort of going into any one specific tool. And then you seem to have kind of the
Tool Training in Data Visualization
00:06:56
Speaker
tool training, the kind of Tableau training or, or Excel training and stuff like that, that's very specific on
00:07:03
Speaker
This is how you build things in this specific tool, but doesn't often tend to include much of the concepts and best practices. What I've tried to do is find a form of training that tries to marry those two together. The tool I'm most comfortable with is Tableau. I'm teaching people how to use Tableau, but in the process of teaching them, I'm also talking about the concepts. If I had to say how it breaks down, I'd say probably it's about 80%.
00:07:26
Speaker
tools, about 20% concepts. But I think one of the nice things about teaching in that way is you actually do get a kind of an overlap, right? Because it's not like it's like, okay, now we're talking about concepts, like as you're, you're doing things. I mean, the example that I use is that I teach people how to make a stacked bar chart in Tableau, like just what you drag where to make it. But then we talk about, you know, is that a good visualization type? And we move the order of the segments and you can start to see, oh, maybe it's not so good. And so
00:07:48
Speaker
How would you convert that in Tableau into a grouped column chart? You can do that just by dragging this here, and then what about a line chart? You can see both compare the different types of visualizations and learn a little bit about what the strengths and weaknesses are of each one, but also you're learning, how do I convert chart types in Tableau in a way that's relatively efficient and clear and allows me to play around with different visualization types efficiently. In a way, I think it's 80-20, but to use a horrible thing to say in data visualization data analysis, I think you almost get 110 percent because there's a 10 percent.
00:08:20
Speaker
So for your journalism students, you're still teaching tableau, but I think this is one of the questions about teaching at a university or teaching more generally, is what tool do you teach?
00:08:35
Speaker
Clearly when someone's bringing you in to teach Tableau, that's what you teach, right? I find this challenging to say the least, right? So like when I teach at Georgetown in the business school, I tend to teach Tableau in Excel, but I wouldn't teach R, not that I could teach R, but I don't teach R. Whereas if I were still teaching in a policy school, R would certainly be on the table and might be something I'd use the entire
00:08:56
Speaker
class the entire way through. And I know, say, at Alberto Cairo's department at Miami, they're teaching JavaScript and D3.
00:09:06
Speaker
stepping away from a moment where you're a tableau is, as you're asked questions about what tools should I use, how do you approach answering that question when people say, well, I can learn tableau now in your class, but when I go and get a job or I go back to my other job, we don't have tableau or we have just Excel. I mean, how do you marry all that together? Yeah. Yeah. I mean, you're right. I think part of it is just what do you feel comfortable doing, right? So I sort of fall into that. Although I do.
00:09:36
Speaker
Think there's advantages to tableau over some other tools the big disadvantage I think is the cost but I think there's advantages in terms of usability I think the big kind of gap in terms of what you teach is I think there's kind of this whole area of
00:09:49
Speaker
coding tools like D3 and JavaScript and things like that. Then there's the non-coding tools. There's Tableau and there's Power BI. I guess R would maybe fit somewhere in between the two, but I think that's the big divide. I think the reality is if you're interested in coding, you're going to go in that direction. If you're not interested in coding or if you've had no experience or exposure to coding, I don't see how you would teach an introductory database class using D3. I only barely understand
00:10:16
Speaker
some coding in Python and stuff myself. It took me forever to get up to just a basic proficiency and even understanding what coding is. So I don't understand how I would teach data visualization and all of that thing about coding basics, and then on top of that, D3 and JavaScript. So to me, if you're going to be teaching non-coders, you're going to be teaching them Tableau or Power BI or something like that. One of the things that I've noticed, Power BI is one of these tools I keep on hearing about and I felt bad that I'd never actually played around with. So I started playing around with it
00:10:44
Speaker
I don't know why I was surprised, but I was surprised.
00:10:47
Speaker
how easy it was to learn given that I knew Tableau. And so I do think in a way, like obviously the tool matters and you want to be teaching the tool that's most relevant to your students. But I've almost got to the point where I sort of feel like what's more important is that you are teaching a tool, right? And that people are actually doing stuff as opposed to just kind of hearing about what the principles are. And then they can then translate the skills they learned in one tool into another tool if and when that time comes.
Comparing Tableau and Power BI
00:11:11
Speaker
Yeah. I mean, I think your point from earlier of marrying the two is really important because
00:11:16
Speaker
Anytime I'm teaching concepts, people want to know, well, how do I actually, you know, what tool can I use to make that stream graph or that treatment or whatever it is? But like you said, you have to know or at least be armed with the skills and the background to understand why you should or should not start a column chart at zero, right? So it is tricky. The Power BI comment you made is really interesting. I've taught some Power BI classes and I sort of come in and say,
00:11:46
Speaker
You just drop and drag, because everybody knows Excel, so everybody knows how to pipe it over from Excel, and you drop and drag, and you're good to go. Let me ask you first, and I'll give you my take. How do you see the differences between Tableau and Power BI? Because I view those as being fairly similar in a lot of instances.
00:12:06
Speaker
Right, I mean I have to say like my the percentages here in terms of how much I've used them like I think I've used Power BI for all of two and a half hours because I don't have it on my Mac so I have to use it on this little PC. Sounds like a jet engine taking off. But I would say that one of the things that struck me they seem very similar in a lot of ways. One of the things that actually found frustrating about Power BI compared to Tableau
00:12:27
Speaker
Power BI has much more the logic I felt of pivot tables where you kind of drag things over to kind of your working area and then the chart gets created, but you can't actually do anything on the chart. So I'm so used to in Tableau when I see a Tableau chart and I want to change the title, I just double click on the title and I change the title or I double click on the axis and I change the axis.
00:12:48
Speaker
reformat something, I right click and click format. Whereas in Power BI, I kept wanting to do that and at least, and again, maybe you can do this and I just didn't come across it, but it seemed like I couldn't interact directly with the charts. I had to do everything on that interaction screen on the right hand, even like titles, like I had to type my title into that little box. I can see why that would be useful from a UI perspective, but I found it frustrating because it felt like I wasn't touching my data in the same way that I feel like I'm touching my data with Tableau. But it's always so hard to know what these tools like you just
00:13:18
Speaker
you've got so much experience with one, is it just that you're having trouble translating that to a new tool? Maybe there's reasons why that's better. It's like when you go from iOS to Android, it's not necessarily that one's worse or better. It's just I'm so used to the way that Android does it or I'm so used to the way that iPhone does it. Yeah, absolutely. No, I think that's right. It's just what you're comfortable in doing. I think one of the things that people have a difficulty wrapping their heads around as they go from an Excel
00:13:42
Speaker
Access Lotus spreadsheet type world to Tableau is moving from that spreadsheet mentality to more of an array mentality. And I think that's the stumbling block. And whereas you get into Power BI, you don't really have to make that switch. But like you said, there are trade-offs of that and what you can do and what you can't do.
00:14:03
Speaker
Interesting. Well, good luck with the online classes. So I want to turn to our second topic where maybe we're going to fight a little bit.
00:14:21
Speaker
Um, so as most listeners know, I just finished my, what I called my month of story where I was talking to a lot of people about story. Um, I wrote a whole bunch of posts about story and I was sort of hoping to get into some fights and you seem the only one, only one willing to get in there and mix it up a little bit. So you wrote a really nice post on news
Approaches to Data Storytelling
00:14:41
Speaker
stories. Let's start this conversation, I think by having you sort of describe your take on, you know, your, your portion of news stories and how it may be different from, from what I was talking about.
00:14:51
Speaker
So I think to keep it nice, we'll start on the part that we agree on, I think, which I think is that I agree with you that for the most part, with some exceptions, thinking about stories
00:15:02
Speaker
like literary stories, these sort of traditional fictional stories with kind of a character and a beginning and an end and a climax and all that kind of stuff is not actually a very effective or useful way to think about data. And that it can either just be frustrating because you're trying to fit your data analysis into this sort of structure that doesn't really fit. It can also I think be dangerous because it can lead you to sort of making facile conclusions that actually your data doesn't support. So what I was sort of making a case for in my blog post was
00:15:30
Speaker
Okay, I think there's another way to think about data storytelling here, which is not as a literary story, but as a news story. And why I think that's more useful is I think a lot of the conventions that journalists use
00:15:42
Speaker
in thinking about story are actually quite applicable to the world of data analysis. So in a news story, you're often thinking about the headline or the lead, like what's the main message here? What's the main takeaway? You're also thinking about kind of a hierarchy of information, like what's the most important thing here? What's the less important thing? And I'm going to structure my story in a way that does that. And you're also frankly thinking about efficiency in a way that you're not so much with literary storytelling, where it's sort of like people need this bit of information very quickly. How can I deliver it to them as quickly?
00:16:10
Speaker
And I also just think fundamentally it's more helpful because in journalism, much like with data analysis, you're trying to inform people. You're trying to teach them something about what's going on. Whereas in literary storytelling, I mean, quite often it's fictional. It's made up. Nothing actually happened. You're trying to kind of entertain or engage your audience.
00:16:28
Speaker
And so, and that's why that sort of structure of literary storytelling is more important. So I was really sort of making a case for why this is a more helpful way to think about data storytelling. I don't actually think it's the way most people do think about data storytelling, but I think data storytelling, I think you correctly identify is this sort of thing that people say without really knowing what they mean. So I guess I'm sort of almost hoping that we can
00:16:46
Speaker
redirect that and say why we say it means this way to think about it as opposed to i feel like there are some people who are pushing things more in the in the literary storytelling i mean i mean cool enough like i mean i i love her book and actually using her book as a textbook in my university of florida course the only chapter that i don't like is the one where she sort of
00:17:07
Speaker
puts this kind of literary storytelling architecture on things because I don't think that's a particularly helpful way always. I mean, it is in some cases, but in most cases, I don't think it's a helpful way to think about data analysis. Right. So let me ask this. So as a journalist, when you're working as a journalist, when you hear the word story, how are you thinking about it? Are you thinking about that sort of inverted pyramid
00:17:32
Speaker
diagram that we all sort of know in this sort of like inform the audience or are you thinking of the sort of more traditional word story. From my perspective, here's what might be a divide in that the sort of analyst type of people that I talk to and work with all the time, I think when they use the word story, they do think of it in that literary sense about the climax and about the drive a new conclusion. And I think
00:17:57
Speaker
Perhaps that's just my own bias and perspective of working in that sort of social science field. And so I wonder, as a journalist, as you're learning and being educated in the journalism school and you're working in journalism, when you hear the word story, maybe you have a different perspective of what that word means to you. Is that fair? Is that right?
00:18:19
Speaker
I mean, it's hard to know what people think when the story breaks and almost like someone to do a survey, right? Because my sort of instinct on it is different than yours. I actually think that, which is actually sort of a point that I think you made early on in the blog post. I think people actually use story quite.
00:18:33
Speaker
loosely or flippantly, right? And I think they sort of use it in a way where they're kind of just, and you see this in other fields too, right? Like public relations or sales. We got to tell a story about our product. What's the story about our product? I don't think they literally mean, you know, where was it born? And who did it fight with in the big climactic scene? They mean, like, we're trying to communicate something, right? And I think, and I think when most people use story, that's what they mean, if they mean anything, and sometimes they don't mean anything is, is they mean something very kind of just very generic that like,
00:19:01
Speaker
And I think what it's almost aspirational, like I want people to care about what I'm saying, right? Like I want to engage people because they're sort of thinking about the stories that they hear. And I'd like my data to kind of tell a
Importance of Audience in Storytelling
00:19:11
Speaker
story. And so it engages people. And so I think actually, they don't really think much more deeply than that. I think it can be helpful to think more deeply than that and maybe use this kind of news storytelling as an architecture of like, this is the way we can think about it, more so than the literary storytelling. I almost feel like one came before the other. I weirdly feel like this sort of data storytelling
00:19:29
Speaker
kind of shorthand, this term that we use, this phrase has been around for a while. I mean, the tapestry conference was arguably kind of created as a data storytelling conference. I feel like it's almost almost in just the last couple of years that people have started to become more literal about it and say like, oh, we mean a story with a beginning, middle and end. I don't feel like that came first.
00:19:46
Speaker
In terms of journalism specifically, I think one of the nice things about story and how it can maybe be applied to data analysis and data visualization is it can mean many things. Fundamentally, a story usually just means an article. It means something I wrote that is going to appear in the paper, it's going to appear on the website, or it's going to appear on the evening news. That can be a traditional narrative feature story that has a character and a climax, all that kind of stuff.
00:20:07
Speaker
It can be a short, brief thing that just includes some important information that people might want to know. It can be like an explainer, which I think is a particularly helpful way of thinking about it from data analysis, where you raise a question to your readers and then you attempt to answer that question. And so I think there's a lot more flexibility there. And I think the way
00:20:24
Speaker
in which it's maybe helpful. I think you could make the case, and I think you sort of made this case, that is this even a helpful way for us to think about data analysis? Maybe we should think about something else. I think it is helpful simply for the fact that the term story makes you think about your audience. It makes you think about who are you telling your story to. And that alone, as basic as that is, I think is really useful because I think
00:20:45
Speaker
So much of data, you can kind of get caught up in the process, like I am analyzing data, and now I'm visualizing the data. And you see this in a lot of organizations where like, people literally see their job is my job is to make reports as opposed to like, well, who are you making the reports for? Like, what's the goal that you're trying to achieve with those reports, right? And so you, it's very easy, I think, in the tools that we use and the data that we're working with to think primarily about process.
00:21:08
Speaker
And so I think simply if talking about storytelling does nothing else, but to get you to sit back and think, well, how is the person that's going to see this chart? How are they going to interpret it? How can I make it engaging for them? It's useful. And then I do think that there's ways in which a news storytelling approach is maybe more relevant and more helpful for people to think more deeply about it than a literary storytelling approach.
00:21:28
Speaker
Yeah, I think that's about right. I mean, I think in your post, you're talking about how people are using the term story in sort of this vague way. And it's not that I, well, I'm not sure. Let me see. I'm not necessarily sure that that's a bad thing, what I want people to do.
00:21:46
Speaker
Or what my hope is that people will use that word carefully. And so maybe I am against the vagueness in some way. I'm not against the inspirational, but when someone says, I have state level data and I want to make a map because the story is that California is high in this metric.
00:22:04
Speaker
I don't want people to say that that's a story, right? And I want them then to stop. And I want there to be another moment where they say, OK, so I don't really mean a story. I just want to show this insight at this point. If I want to tell a story, then I need to do this other thing. And so maybe being vague at the beginning is OK, but then I'm hoping people will hone in, right?
00:22:32
Speaker
So yeah, I'm not sure whether the being vague is a good or a bad thing necessarily, but I do. I think my whole point of the whole month was really to get hopefully to get more people thinking about their use of that term, right? Well, I guess what I would say is that and I think I tried to make this point in the blog post.
00:22:53
Speaker
The problem I think you and I agree on is that being very specific and thinking about data storytelling as literary storytelling is not very effective in most cases, could be dangerous in some other cases. We don't want our field to go down this route where people are constantly thinking, I have to tell a literary story and what's my climax and who's my character? Because I think that's going to cause a lot of forms of data analysis and visualization to go off the rails. I guess what I was arguing is that the other problem you identified, which is that people are using this term flippantly or vaguely,
00:23:22
Speaker
in some ways gives me comfort.
00:23:25
Speaker
That that means the other problem isn't as bad because if everyone meant literary storytelling I'd be much more worried because I don't think that's very effective the fact that people are kind of using it in this kind of vague way I sort of feel like that actually gives us an opportunity to Define it in more helpful ways or to steer the conversation in more helpful ways And I guess I mean in some ways some of this is semantics, right? Like what are we calling this thing? What are we calling this thing? I mean and I'm kind of a bit of a fatalist where I sort of feel like you know Like people don't like it when people use the word literally when they mean metaphorically, but but they do so
00:23:54
Speaker
It's too late. We can't get people to stop using the term to do the storytelling because it's done.
Core Elements of Storytelling
00:24:01
Speaker
I think a lot of it is semantics. There are lots of definitions of what story is and what it means and we could go look in the dictionary about the word story and there's just something about
00:24:16
Speaker
those definitions that didn't, for me, just didn't capture what the essence of what it means to tell a story or to hear a story. And so I tried to come up with my own definition of these two criteria of having emotion, having real emotion and have a meaningful climax and whether that helps people think about how they would present their data. I don't know if that helps or not, but it at least
00:24:44
Speaker
provide some concrete characteristics of that word that I think can help get people started in thinking about it. I do like this idea that
00:24:54
Speaker
having people who are working with data think aspirationally because we know things like snowfall and the piece from the Tampa Bay Tribune that I'm forgetting the name of the failure factories and the piece that the Washington Post immigrants are on the border that I'm now forgetting the name of too. I mean, we know those are successful and lots of people read them and lots of people share them and we know that
00:25:21
Speaker
that happens because at their core they are stories in this, I don't know, I don't even know the traditional sense, right? It's a news story, but only insofar as that it's a news organization publishing it.
00:25:35
Speaker
Yeah. Right. But at its core, it is a story. And so I think that being an aspiration for people who are making a starting with a bar chart is inherently good. If we all knew how to marry data and photographs and traditional storytelling, I think we would, well, we'd all be saturated, but we would, you know, we would be able to marry those things together and really drive our analysis to many more people and get them to care a lot more about them.
00:26:02
Speaker
Yeah, it's interesting. I mean, because in some ways, I think I'm actually more down on story than you are, because I sort of look at it and I think there are opportunities to tell these more epic kind of stories that have data at their core. But I guess I sort of wonder, like, in how many situations is that actually true? Is that actually the best route to go? I tend to think it's going to be a pretty small, like single digit percentages, right? And I use this sort of example,
00:26:27
Speaker
I would sometimes joke that in journalism we have what I call anecdote-itis where it's basically like we always feel like there needs to be an anecdote and so you'd be writing a data story and you feel like okay I need to have this character that's going to be sort of opening the story and then I'll have my nut graph with the data in it and stuff like that. But I sort of compare it to like if you open up your smartphone and you're wanting to know the weather and you go to your weather app and instead of just seeing the temperature and the forecast for that day
00:26:48
Speaker
you started telling you a story about some woman who woke up that morning and she's hoping to go golfing, but she can't go golfing if it's going to rain that day and the clouds are coming in off the Pacific. You'd be like, shut up. I just want the information. In a lot of these cases, I really feel like
00:27:03
Speaker
It's not helpful to think about things in terms of traditional literary stories. And the reason why stories are so pervasive in entertainment is because it's made up, right? Like there's no information to impart. So the whole purpose of it is to kind of draw you into this fictional world and make you care about characters who are fictional. That's not the case with
00:27:22
Speaker
with information, whether it's news or whether it's data. If you want to know if bombs were just dropped on Syria, a simple one-line news alert from the New York Times might be the most effective way to tell you that information. In a data analysis perspective, a simple, clear dashboard that lets you see at a moment what sales are like in your organization may be the most effective way to present that.
00:27:43
Speaker
Yeah there are places where we can tell these more epic stories but i feel like it's a bit of a trap to sort of feel like every time i do a presentation every time i do a visualization i need to tell a story and i need to kind of slowly build to a climax because i think in a lot of cases that's not gonna be the most effective way to inform your audience in the most effective and efficient
Simplicity in Data Visualizations
00:27:59
Speaker
way possible.
00:27:59
Speaker
No, I totally agree with that. And I think I would extend it a little bit to even say that we often make interactive visualizations when really we don't always need an interactive. And there's been a bunch of blog posts on that that people are debating the sort of useful utility of interactivity. But you see a lot of people are like, oh, I'm going to make this interactive visualization where
00:28:22
Speaker
I don't know, hovering on the 275 lines with 40 numbers on each line doesn't really give you anything except the opportunity to write down 9 million numbers, which is not really helping you. Yeah. So yeah, I think there's a lot of places where
00:28:41
Speaker
maybe it's sort of the pull back a little bit or I don't know, like data visualization retro, right? To go back to like, let's just make a bar chart and call it a day, right? And that's, and we don't need story. We don't need interactivity. We just need the simple bar chart. But I mean, I've seen that in Tableau in particular, and it's interesting because a lot of the Tableau kind of
00:29:01
Speaker
example dashboards that they show you are an example of what I kind of jokingly call, you know, show everything, filter everything by everything. It's just kind of barfing up of data with no kind of, to use a news term, an editorial judgment about this is what's important, you know, we should focus on this. I think partly that's often
00:29:19
Speaker
uh, caused by the fact that we're sort of worried because we're producing a visualization for a boss and we're worried. The last thing we want them to say is, well, why did you leave this out? So there's always this bias towards we'll leave everything in and, and these tools allow us to leave everything in because, because they can digest huge quantities of data and we can have all, all these sort of filters and stuff. And so I think a lot of it is just kind of having the discipline to say, no, like this is what's important.
00:29:40
Speaker
So either I'm not going to have any interactivity at all. I know that this one simple bar chart is what tells the story. And so I'm going to focus on that. Or I should have a limited amount of interactivity, but I'm only going to let people filter these two or three things because that's what's important. So again, I think it's this sort of it's constantly thinking about.
00:29:56
Speaker
Who's the audience? What do they need? What's going to help them?
Conclusion and Future Discussions
00:30:00
Speaker
And again, that's why I feel like if nothing else, thinking in terms of story gets us to sort of change our perspective a little bit, as opposed to kind of getting stuck in the numbers and just feeling like, well, I visualized the data, so I've done my job. But have you done it in a way that's effective and helpful to your audience? Yeah. Yeah. All right. Well, on that note, I don't think we fought enough. So I'll have to end.
00:30:22
Speaker
You're going to have to come back on and we'll have to have another fight. But yeah, Chad, thanks so much for coming on the show. This has been a really fun conversation. Thanks a lot for having me. And thanks to everyone for tuning in. Chad and I talked about a lot today and there'll be a long list of resources and sites for you to check out on the show notes page. So please do check that out. Thanks again for tuning into this week's episode. So until next time, this has been the Policy Vis Podcast. Thanks so much for listening.