Hey Meg, let's do a podcast.

That's a great idea. What should we talk about?

So, Meg, last year the book data and analytics for instructional designers came out, and one of the things that people have said was like, wow, we thought this was going to be about charts and graphs. And Megan, you've only got one chapter in an entire book that talks about data visualization, and yet I don't want us to think that that means it's not important. Visualization, it's a big word. It's easy to flub. We've flubbed it, what, four times already before this version of the podcast. Maybe we'll make it through on this one. Let's just call it Dataviz.

That's a good idea.

Okay, why should, despite the scant treatment received in the book, why should learning and development people care about datavizability?

Great question. And you don't have to take it from me or from you. Data and analytics is one of the eight capabilities of ATD's talent development capability model. That's also a mouthful. And under data and analytics, there's listed several skill statements that are related to Dataviz. There's knowledge of data visualization. See, I still can't do it. Data knowledge of data viz.

Including principles, methods, types and applications, for example, texture and color mapping, data representation, graphs and word clouds. So the list goes on on how Dataviz relates to this capability in ATD's model. Now, as you mentioned, you know, you have one chapter, it's a one piece. It's one piece of what we have to do as lnd. And because of that, you know, we all have teams of people. Hopefully we have teams of people or we live, we work in an organization that has other people that have skills that may not be our strength. So I don't want to scare anybody saying like, okay, everybody gotta go out there and take stats 101 and get up to speed on all the numbers and all of that. Because, you know, I know that's not everybody's jam, but it is something that's important because the things that we're creating as instructional designers or as L and D teams, they're going to be creating data.

And that data will be important to our organizations and can help us tell a story. So, Dataviz, in the end, it's really all about communicating what's happening. It's communicating relationships between different pieces and parts of what's going on with our courses and with our learners.

It's almost like a picture is worth a thousand words or a viz is worth a thousand rows in an Excel chart.

Yes, yes, absolutely.

So given that, right, and you mentioned other people in the organization. So, like the folks in finance or the folks in marketing, or like, there's other people around who have to do the same. So you may not be, you may be the only L and D person in your organization or not, but there's other people in your company who can do this, too. But if an L and D person wanted to get started, like, what are some good places to go, since they obviously can't go to my book to get much on data visualization.

Right. So there's two main people that I've heard a lot about when I've started to just search the Internet for more about Dataviz. So the first person that pops up if you search for stuff is Edward Tufti. He's a statistician.

Another hard word.

Another hard word.

Stats guy.

He's a stats guy and a professor. And I went to his website recently, and it said the quote from somebody else said he's the da Vinci of data. He's got five or six books, one of which is called the visual display of quantitative information. And he's got some amazing stuff in his books. I've got several of them at home. But a lot of his Dataviz stuff is not necessarily business related. So he's got one. That's the path that Napoleon took through Europe or something along those lines, and a way to represent that visually.

So probably not as relatable to us in the L and D and business world. But another person that has some really great stuff out there is Steven few. And as I was looking at his book recently, he was talking about how what he's really passionate about is education and learning and helping people learn and understand more through Dataviz. And so I've. And years ago, I went to his website. He's got some awesome stuff on there about examples that really clearly illustrate what not to do when creating a dataviz. So things like, you don't want to turn something 3d, like a chart or a 3d pie chart, really not a good idea. And the examples, because he's showing them to you, you can really start to see and understand why one helps tell a story better than the other.

Because we have visual limitations. Our eyes can only understand area in 2d, not necessarily 3d. It kind of messes with our perception of quantity. So he's got a great book. It's called show me the numbers. And because his focus is a little bit different than Edward Tufti. His is more geared towards the masses, helping everybody have a better understanding of Dataviz and business specifically. Just a warning, though, if you go to either of their websites.

These gentlemen have been in this business for a very long time. Their websites look like they were built in maybe the late 20th century, partly because they've been around that long, but also their focus is on simplicity, because Dataviz really is done best when you keep it very, very simple.

That's funny. You said late 20th century, and all of a sudden I'm doing some math. I was like, oh, my gosh, we are a quarter of the way through the 21st century.

True. It's true.

Right? But some of these are timeless principles, and I think that's important to understand here. And I'll actually throw two other, if I can, into the mix and learning space. You and I went years and years ago to Tom Crawford's vizlet conference. He ran, I think, three or four of them in Michigan, and it was super powerful and in entire day, just drawing and communicating visually. And it was data, but we covered a lot of other topics. Those were fascinating. And then Kevin Thorne actually is doing a fair amount of that work right now. And we can link to that in the show notes, but he's talking about, how do we communicate visually? Using simple shapes that everybody can use.

And so that simple shapes that he's talking about gets me back to what you were saying, which are 3d shapes, because we read too much into a 3d shape, but they look so fancy and we can make them in Excel, and they can be, like, shaded and shadowed in three D. And. Yeah, it's just. It actually makes it harder to read the information.

It does. It's like extra visual clutter that your brain has to filter out for you to then interpret what it actually is trying to tell you.

Wait a minute. That sounds like an instructional designer talking. Are you talking about cognitive load?

Maybe a little bit.

All right, see, I think that once you get past a certain amount of math, apprehensive apprehension, more big words that are hard for us to say that because the average learning and development person is not in learning development because they love statistics and math.

Yeah, I think you're right about that.

Right. Okay, so given that, though, when it gets down to, I mean, you've talked about communication, storytelling, taking complex subjects and making them easy and reducing cognitive load and working with the brain. So actually, all of those things are things that learning and development people do all the time.

Yep.

All right, so let's just get over the math. Okay.

Yeah, we can do that.

Okay. So, um, we've kind of gone off on a lot of tangents here, and that's okay.

That's the name of this podcast, Megan.

Ah, see, we were pretty smart there. Yeah. Past us was pretty smart. What? And I just, while I've got you here, we will probably come back and talk about data visualization in other podcasts. What are some of your. Just like, quick tips?

Yeah. So definitely, number one is don't ever use like three D bar chart, 3d pie chart. They're not readable. In the end, just keep everything 2d. There's no need or reason to go that direction. I think I've even seen examples where people try to dress it up with putting a picture, an image behind a bar chart. There's no need for that either. That's just distracting from the data itself.

And that kind of is my next point is like, let the data speak for itself. Like, don't add more labels, add more images, add more shadows, any of that sort of stuff. Keep it very, very simple. And then there was one more, which I'm forgetting at the moment, so we can cut this little piece out.

Oh, no, no, no, we can't, because you were talking about paying really close attention to where the X and Y axes intersect.

The one I was trying to remember. Yeah, sure. So when you're creating, like, any kind of chart that's on an xy axis, you always, always, always need to keep the bottom of that chart at zero. Because if you pull that up, like, let's say you pull it up to 400 and your numbers are 401 and 463, and then that messes, and it makes it look like 463 is so much huger than 401. And so there's some accuracy, things that are really important to keep in mind.

So I think that's a huge point. Right? Because when you have a graph, sometimes it looks like, oh, that's very smart. You brought the data. And it can be sometimes easy, I'm going to say, to mislead, as though that's intentional. And sometimes it is, but also just unintentionally to be telling the wrong story with data. But it's easy to miss that because it's a pretty shortened graph, right?

Yeah. Yeah. And, you know, you talk about misleading. A lot of the examples that you'll find on Stephen Fus website are things that maybe a particular person has put out in the world because they have a story to tell. They want you to believe that XYZ is happening out there in the world. And so they'll create a dataviz that seems to be telling a certain story, but actually they've sort of messed with how that data is represented visually so that it better supports the message that they're trying to tell. And so I think that's really important is, like, we have to be very, very clear and vigilant and sure that we know that the data is behind the visualizations that we're creating, so that we're not intentionally or unintentionally putting out into the world something that could be misinterpreted and create falsehoods out there, or have people come to a conclusion, and then they make a decision based off of that data visualization. And maybe it wasn't quite right.

So you need to keep in mind things like your sample size, like, how much data is this really based on? Like, is it ten pieces of data or thousands of pieces of data that we're looking at?

So basically just good responsibility with numbers.

Good responsibility. So important. Yep.

So how'd that go, Meg?

I think it went pretty well. You know, we got some words that were hard to pronounce there.

Yeah, big words.

Big words get stuck in your mouth when you're a human.

We're not AI.

We are not AI. This is proof. And I guess the other thing is, you know, it's really hard to talk about data visualization, something that is visual on a podcast where it's just your voice.

Yeah, I kept wanting to see pictures. Pictures, yeah. Okay.

Maybe we can include those in our show notes or on our website.

There you go.

All right, sounds great.

This is Megan Torrance with Meg Fairchild, and this has been a podcast by Torrance Learning.

Tangents is the official podcast of Torrance learning, as though we have an unofficial one. Tangents is hosted by Meg Fairchild and Megan Torrance. It's produced by Dean Castile and Meg Fairchild, engineered and edited by Dean Castile with original music also by Dean Castile. This episode was fact checked by Meg Fairchild.