Join Bill Shander for an in-depth discussion in this video Tools, part of Data Visualization for Data Analysts.
- You're a data analyst. You have your tools. You're probably working in SPSS or STATA or R or whatever. You're working in tools that are built for your data, right? But you may or may not be working in tools built to make beautiful and simple graphic displays of that data. In this movie, I'm gonna talk briefly about tools and how to use them and think about them as well as to try to introduce you to some tools and functionality that you may or may not be aware of. First, let's talk about the standbys that you're likely using. Whether you're in Excel, STATA, SPSS, or another tool, you can and probably already do generate your own charts and graphs.
This is what you do all day long, right? Okay. But here's the thing, all of those tools, by default, don't follow the rules I'm laying out for how to design effective charts and graphs. The default colors tend to be overwhelming. The axis are over-emphasized, etc. However, in most, if not all of these tools, you can format your charts to do what I'm suggesting. And some of these tools have the ability to load in styles or templates. So if the tool you use can do this, I recommend it. Just create a chart, apply the styles and formats that work according to best practices, save those settings and then you can apply that style to future charts of that type.
So don't tell me your tools can't do what I'm recommending, the odds are against it. That being said, every software tool is limited in what it can produce, unless you're coding from scratch. You're always going to be trading flexibility for speed when using tools like this. So let's assume you want to push your designs a bit further or you want to use a chart type that's not available in your tool of choice. Many data analysts won't abandon their tool. Conversation over. I'm not listening (laughs), right? But if you're willing to consider it, there's a huge selection of tools available to you to work with.
So these two websites, selection.datavisualization.ch and visualisingdata.com/index.php/resources, you can see the urls here. These are the two websites that I'll go to look for tools when I'm thinking about experimenting beyond my normal tool set. I usually start at this one, datavisualization.ch because I've been using it longer and it's smaller and it's simpler and then if I don't find something here, I'll check out the other one. So let's talk about the various options here. I think of these tools that we're examining here in different categories.
So the website actually allows me to filter them. So I can look at just tools that have to do with mapping, right? Or just charting tools or data tools. And this other website has even more filtering, right? I can look at just things that have to do with programming or multivariate chart typing tools, specialist tools, color-related things, etc. These filters make it a lot easier to find the tools that I might be looking for. But what I really like about these websites is that if I go back to All here, let's say I wanted to do a really alternative chart form like this one.
If I'm gonna click into here, I can get some information on what it is and, of course, a link to the website for this tool, but you can see, you know, this type of chart form, Excel is not gonna ever be able to generate something like that, right? Or if I want to create a network graph, quickly and easily, again Excel is not so good at doing this. You can't really do this, Gephi is another tool that'll do that. As I mentioned before, mapping tools, lots of mapping tools in here. All kinds of these alternative chart forms that I really can't do in my standard tools. And as I mentioned, if I want to find programming tools, right, so in here I will find things like I've been mentioning: D3.js, other opensource charting libraries, like chart.js.
It's not the only one, it's just my favorite. Now I can't talk about tools without talking about the standbys on the design side. Let's say you're working in Excel and you create a bar chart and you have a decent template with your corporate colors and your colors are pretty good but you want to do something a bit different with the chart. You want to add a call out of some kind and maybe you want to colorize one bar in a different color and make it pop in a certain way and maybe add a drop shadow, whatever it is. So you want to do some custom design on top of a standard chart form. What I would recommend is, again, if you're up for it, is to get a copy of Adobe Illustrator.
So Adobe Illustrator is a vector graphics program. And you can create graphics in here that are usable on the web, as well as in print, or for really any other use. And what's really great is that you can copy and paste charts right out of Excel and into Illustrator. So I'm just gonna do that really quick for you here. So if I click on the chart-- so I've created this thing, it's a standard Excel chart. I just click on it and copy it and now if I go back to Illustrator, and if I paste it you might get some weird errors, you can usually just ignore those, it comes in as vector graphics.
Now, look what happened here, my fonts are all crazy. So you might sort of freak out for a second, that's okay. I'm just gonna delete that for a second and go back to Excel. What that is is Excel for some reason, again, example of the standard form being a little bit odd uses a default font that Illustrator doesn't like, so if I just right click on my chart, say Format Chart Area and change the font, instead of from using Calibri, maybe I can use a more standard font like Arial? That should do the trick. Now if I copy and paste that into Illustrator, it's gonna be beautiful and perfect.
And so this is vector art, as I said, which means I can scale it up, I can scale it down endlessly. And now I can go in here and I can really work with this. So I can, let's say, take the title off, okay? And I can take one of my bars and do different things with it. Or I can remove a bar if I don't like it. Now it does some strange things with how it handles shadows in Excel sometimes, but I can really get in here and mess around with my chart quite a bit. I can remove the axis, I can remove my titles, I can replace them with different ones, I can add graphics on top of this. I can really do anything I want with this, but I have a data-accurate bar chart that I can really customize to my heart's content.
And by the way, of course, if you're not comfortable working in Illustrator, you don't have time to do this, it's good to at least know that this can be done, meaning that you can hand off data-accurate charts to your designers to work with and they can do what they need to do with really, truly accurate visuals. We could probably do a whole course on tools for data visualization. As you saw on this website, there's a nearly endless number of tools available. I recommend that you look at these tools, experiment with them, learn at least one or two. It won't take you long to pick up the basics, which will allow you to produce more compelling visuals to showcase your compelling analysis.
- Why visual communications matter, and how they work
- Communicating via story
- Communicating with color
- Using legends and sources
- Sketching and wireframing
- Rethinking slides, charts, and diagrams
- Rethinking your templates and brand guidelines